Content uploaded by Željko Poljak
Author content
All content in this area was uploaded by Željko Poljak on Mar 05, 2024
Content may be subject to copyright.
1
Give the media what they need: Negativity as a media access tool for politicians
Željko Poljak
University of Antwerp
zeljko.poljak@uantwerpen.be
Forthcoming in The International Journal of Press/Politics
DOI: https://doi.org/10.1177/19401612241234861
Abstract
Recent studies indicate that politicians' negativity usage fails to enhance their approval ratings
among the general public, yet politicians regularly use it. This begs the following question: why
are politicians so negative if this strategy does not bolster their prospects for re-election? In this
paper, I argue that the media, driven by audience engagement, plays a pivotal role in shaping
politicians' propensity for negativity. Specifically, politicians resort to negativity because it aligns
with the media's negativity bias, thereby increasing their chances of securing media access and
public attention. I test this expectation on the less-likely case of Belgium, using data on politicians’
negativity usage in parliament and their presence in prime-time TV news (2010-2020). The results
show that using negativity significantly increases politicians' chances of gaining media access,
particularly when using uncivil negativity. The more media access politicians start to attract due
to negativity, the more they resort to negativity.
Keywords: Incivility; Politicians; Mediatization; Media Logic; Negativity
2
Introduction
Negativity in politics is a multi-layered concept (Lipsitz and Geer 2017). For instance, it can relate
to media outlets reporting political news with a negative tone and a pessimistic outlook (Lengauer
et al. 2011), or to citizens actively engaging with negative political information (Soroka 2014).
Concerning the primary suppliers of political negativity, namely politicians, negativity is
predominantly defined as "any criticism leveled by one candidate against another during a
campaign" (Geer 2006: 23). Occasionally, this type of negativity may deviate from social norms
(Mutz 2015), resulting in uncivil exchanges as politicians engage in mocking and name-calling
(Sobieraj and Berry 2011).
Such negative and uncivil communication between politicians can have a profound impact
on society. On the one hand, negativity can positively affect the public by increasing their political
attention (Mueller and Saeltzer 2022; Mutz 2015; Soroka 2014) and satisfaction with democracy
(Ridge 2022; Tuttnauer 2022; Van Elsas and Fiselier 2023). On the other hand, it can also yield
adverse consequences, such as lower voter turnout (Nai 2013), heightened public anger (Gervais
2017; Walter and Ridout 2021), increased resentment toward politics (Van't Riet & Van
Stekelenburg 2022), and the amplification of affective polarization (Druckman et al. 2019; Iyengar
et al. 2012; Skytte 2022). Yet, while we know a lot about the impact of negative political
communication between politicians, it is imperative to elucidate the underlying causes that drive
politicians to be negative. This is especially vital in light of the escalating and pervasive nature of
negativity and incivility in political discourse (Frimer et al. 2023; Geer 2012; Maisel 2012).
Scholars who study the causes of negativity between politicians often argue that negative
rhetoric is used to increase one's electoral attractiveness at the expense of political opponents
3
(Harrington and Hess 1996; Skaperdas and Grofman 1995). However, this strategy does not
always pay off. The so-called “boomerang effect” of negativity can lead to declining approval
(Fridkin and Kenney 2004; Nai and Maier 2021) and may increase voters’ preference for other
candidates (Somer-Topcu and Weitzel 2022; Walter and Van der Eijk 2019). This makes the usage
of negativity by politicians a risky strategy, the outcomes of which are often uncertain (Lau et al.
2007).
Despite these uncertainties about the effects of negativity, some politicians believe that
discrediting and attacking opponents is their main job (Sevenans et al. 2015). Studies have shown
that this is indeed the case: criticism among politicians is not something that only occurs during
hostile campaigns, but in regular day-to-day politics as well (Ketelaars 2019). As such, based on
these findings, one might get the impression that the rationale of politicians regarding negativity
usage is that the benefits (i.e. more support among the voters) must outweigh the costs (i.e. less
support). But this is far from true. A recent longitudinal and comparative study shows that
politicians’ usage of negativity in day-to-day politics does not lead to increasing public approval
in public opinion polls, and more often than not results in decreasing approval (Poljak 2023b). But
then why do politicians persistently rely on negativity by criticising each other if they see that this
does not benefit their voter support? This question is particularly puzzling knowing that politicians
admit that their approval among the public is an important indicator in their daily work (Oleskog
Tryggvason 2020).
In this paper, I argue that the media's negativity bias, fuelled by the audience's
engagement with negative news, may explain why politicians continue to use negativity despite
its uncertain impact on public approval and negative side effects on society. While I am not the
4
first to theorise about the relationship between political negativity and the media (see Ridout and
Walter 2015), previous studies focused exclusively on short-term campaigning periods
establishing that negativity usage leads to media coverage (Haselmayer et al. 2019). While these
findings are fundamental, they do not explain the persistent usage of negativity by politicians
beyond campaigns, nor do they tell us whether media coverage of negative politicians reinforces
negativity usage.
To clarify the interaction between political negativity and media, I rely on the literature
on the mediatization of politics (Strömbäck 2008). This theory argues that politicians adapt their
behaviour to media logic as a means of achieving media access. Politicians need to do this as
gaining continuous access to the media can significantly benefit their re-election goals (Van Erkel
and Thijssen 2016; Van Remoortere et al. 2023). However, it remains unclear which media logic
politicians should follow to reach the news and several studies offer different arguments (see
review in Vos 2014). Nevertheless, we do know that media logic embodies techniques media use
to capture citizens’ interest (Strömbäck 2008). Given the public's tendency to react more strongly
to negative information compared to positive ones, media outlets often operate under a
negativity bias (Soroka 2014). Consequently, I anticipate that politicians cater to the media's
demand by providing them with what they need to engage their audience: negativity.
In this paper, I formulate three hypotheses: (i) negative politicians have a higher chance
of appearing in the news, (ii) especially if they use uncivil negativity and, (iii) once politicians gain
media access following negativity, they are incentivised to go negative again. These hypotheses
are tested on the less-likely case of Belgium, using data on negativity usage by 367 politicians in
the country’s federal parliament during an 11-year period (January 2010 - December 2020). More
5
specifically, I analyse whether a politician goes negative during question time sessions in the
parliament at noon (t-1) and regress this decision on a possible news appearance later that day
during prime-time TV news at 7 pm (t). In turn, I also test whether appearance in the news due
to negativity (t), leads politicians to go negative again (t+1). The results significantly confirm the
expectations. Negative politicians in parliament have a higher probability of getting into prime-
time TV news, especially if they use negativity that is uncivil, and once they gain media access
following negativity, this experience increases the probability of going negative again. As such,
the results confirm that politicians try to adapt to media logic, as they are likely aware that using
negativity results in media access and public attention.
Negative politicians
The literature studying politicians’ negativity usage has already explored the causal mechanisms
behind the usage of negativity linking it to a vote-seeking political strategy (see review in Nai and
Walter 2015). Negativity is used as a means to diminish the electoral attractiveness of opponents,
and increase one’s own chances of election victory (see also functional theory by Benoit 1999).
While recent studies challenge this notion by showing that certain personality traits also
contribute to higher negativity usage (Nai et al. 2022), the general principle of politicians’ rational
approach to using negativity holds: politicians who consider the use of negativity to be a beneficial
rather than costly strategy, are more likely to go negative (Maier et al. 2022). This can be observed
by looking at how public approval impacts politicians’ negativity usage. Politicians with low
approval ratings are more likely to resort to negativity compared to politicians with high approval
6
ratings (Damore 2002; Elmelund-Præstekær 2010; Haynes and Rhine 1998; Maier and Jansen
2017; Nai and Sciarini 2018; Walter and Van der Brug 2013).
However, while negativity can potentially increase approval ratings, or at the very least
diminish the approval rating of the competition, there is a boomerang effect associated with
negativity usage. Nai and Maier (2021), for instance, demonstrated that harsh negativity does not
diminish the evaluation of the target but rather reflects negatively on the one employing
negativity. Furthermore, a study by Walter and Van der Eijk (2019) points out that using negativity
may lead to a second preference boost, whereby citizens may choose to vote for their second
preferred candidate if their initial candidate goes negative. In addition to that, targets of attacks
may receive higher voter support as opposed to attackers (Somer-Topcu and Weitzel 2022).
Recently, a longitudinal and comparative study from Europe showed that politicians’ negativity
usage hardly impacts public approval in opinion polls, and when it does, it is most damaging for
the attacker (Poljak 2023b). These effects are especially applicable in multi-party systems where
voters have multiple ideological proximate candidates to support (see Walter 2014; Walter et al.
2014).
Still, negativity remains widespread and present in a day-to-day context (Ketelaars 2019).
This begs the question: if politicians see that their negativity usage does not impact or even hurt
their approval, why do they continue using it? We could simply assume that politicians remain
unbothered by, and do not care about public opinion polls. But this notion contradicts many
studies on political representation, which have shown that politicians tend to follow public
opinion daily (Wlezien 1995) and admit that their approval impacts their day-to-day work
(Oleskog Tryggvason 2020). After all, public approval polls are the most accurate predictor of the
7
election outcome, especially closer to elections (Jennings and Wlezien 2018). Therefore,
politicians are affected by public approval, and it remains unclear why negativity remains
continuously used even though it is hardly paying off (Poljak 2023b).
To disentangle this phenomenon, I argue that we need to place greater focus on the
media, driven by audience engagement, as a predictor of negativity usage. As stated before, I am
not the first to explore the relationship between political negativity and the media. Geer (2012),
for instance, argued that negativity in politics is rising as a result of increased negativity in media.
Following this premise, Ridout and Walter (2015) have theorised that politicians supply media
with negativity as this strategy has more chances of getting picked up by journalists. Although
their analysis of political advertisements in the news did not identify a strong bias towards
negative political ads, Benoit et al. (2004) and Haselmayer et al. (2019), for example, did show
this to be the case. Among campaign messages politicians supply to the public, negative messages
are more likely to reach the news as opposed to non-negative ones. Furthermore, politicians who
employ more negativity tend to receive greater media coverage (Maier and Nai 2020), and
politicians in general are more likely to be featured in negative news stories (Niven 2001;
Vliegenthart et al. 2011).
Despite these fundamental findings, the (short-term) campaign-centric perspective does
not tell us anything about the media coverage of negativity in routine politics. It is crucial to move
beyond campaigns because it is likely that the daily presence of negativity contributes to citizens'
satisfaction with democracy (e.g., Tuttnauer 2022), but also growing resentment towards politics.
This resentment erodes political trust (e.g., Thorson et al. 2000), reduces voter turnout in
elections (e.g., Nai 2013), and exacerbates affective polarization (e.g., Iyengar et al. 2012).
8
Moreover, previous studies investigating the relationship between politicians' use of negativity
and news media primarily treated the media as a dependent variable, overlooking the exploration
of whether media coverage reinforces negativity among politicians. To the best of my knowledge,
it remains unknown whether media coverage of negative politicians amplifies the inclination for
negativity within these same politicians (but do see Geer 2012).
Mediatization of (negative) politicians
In the era of social media, politicians have unprecedented opportunities to engage with the
public. Yet, while social media has considerable influence on the public (Conway et al. 2015),
traditional media continue to play a vital role in politics (Djerf-Pierre & Shehata 2017). As a result,
politicians still consider traditional media more pivotal for informing citizens about their activities
(Soontjens 2021) and for obtaining feedback from the public (Walgrave and Soontjens 2023)
when compared to social media. For instance, while social media primarily enable politicians to
connect with their existing followers (Jürgens 2011; Peeters 2022), it does not reach the general
public that consumes media sources beyond social networks (see also Dubois and Blank 2018).
Therefore, it is reasonable to assume that politicians continue to rely on traditional media access
to communicate with the public.
Two broad goals can be on politicians’ minds when they try to gain media access: personal-
driven goals or policy-driven goals. Personal goals relate to politicians being occupied with their
own image to attract votes at elections and advance their careers. For example, recent studies
show that having more media access increases vote share at elections (Van Erkel and Thijssen
2016). Furthermore, being in the news allows politicians to climb the ladder in their careers by
9
reaching positions of party leaders, ministers, or prime ministers (Van Remoortere et al. 2023).
Politicians themselves admit that media can make or break them (Van Aelst et al. 2008). This is of
particular importance with the increasing partisan dealignment where voters become volatile
(Dassonneville 2023) and base their vote choice on individual politicians, rather than parties
(Garzia et al. 2022).
While certain politicians may be primarily driven by their personal goals, politicians may
also want to gain media access to advocate their policies. Instead of paying attention to their
public image (see Louwerse and Otjes 2016), certain politicians see their main role as policy
advocates that genuinely care about tackling issues and implementing their ideology and policy
solutions (Sevenans et al. 2015). Pushing for an issue through media coverage can be particularly
useful due to the agenda-setting power that the media has over political agendas (Vliegenthart
et al. 2016) and politicians are well aware of this fact (Walgrave 2008). Therefore, if a politician is
successful in gaining media attention, their issue and policy solutions receive a spotlight as well.
Ultimately, regardless of when we observe politicians or what their underlying goal is, it is likely
to expect that vast majority of them crave media attention.
While politicians’ dependence on the media cannot be understated, gaining media access
is not an easy task. Journalists and media editors act as gatekeepers determining who gets into
the news (Shoemaker et al. 2008). For example, a journalist is more likely to report on stories that
are going to generate greater sales or increase viewership and visits to the media webpage.
Journalism studies on metrics nicely depict this phenomenon. The more citizens click on the news
and increase the metric of the news story, the more journalists tend to prioritise these stories
10
(Lamot and Van Aelst 2019). Therefore, politicians that want to increase their chances of getting
into the news need to adapt to media and journalists’ preferences for what is relevant.
This phenomenon is the crux of the literature on the mediatization of politics which claims
that politicians adopt the media logic (Strömbäck 2008). Media logic represents the
abovementioned techniques the media uses to capture citizens’ attention which is visible through
news values that make certain stories newsworthy such as conflict (stories about arguments or
controversies), entertainment (stories that are humorous) or shareability (stories that generate
engagement) (Harcup and O'Neill 2017). Therefore, politicians need to use media logic by
following these news values to attract journalists’ interest, that is, they need to provide journalists
with something controversial, funny, etc. Indeed, empirical applications of the theory highlight
that politicians do abide by the media logic trying to use it for their own advantage (Elmelund-
Præstekær et al. 2011). Yet, studies offer different reasons about the exact type of media logic
and news values that allow politicians to enter the media (see literature review in Vos 2014).
This is where the negativity bias comes in (Soroka 2012). This concept may help us to
disentangle politicians’ negativity usage as a media access tool. Negativity bias in the news is the
result of human psychology: we attribute more importance and heavier weight to negative, rather
than positive information (Knobloch-Westerwick et al. 2020; Soroka and McAdams 2015; Soroka
et al. 2019; for an additional cultural perspective see Shoemaker 1996). Therefore, as consumers
of news, we are more prone to consume negative rather than positive news (Fournier et al. 2020;
Robertson et al. 2023). This nudge from the public reinforces journalists’ media logic to use their
gatekeeping role by providing negativity as this generates more attention for the news outlet. As
such, negativity bias in the news can be seen as a concept that reflects media logic relying on
11
news values such as conflict between two sides (Lengauer et al. 2011), driven by audience
engagement with such attractive news. Even in situations where the negative news does not
attract attention (Muddiman et al. 2020), journalists’ gut feeling or practices will lead them to
rank negative news higher (Bartholomé et al. 2015; Lamot and Van Aelst 2019).
Accordingly, if politicians want to gain media access, they need to provide the media with
what the media needs: negativity. However, if the majority of politicians adopt such media logic
by using negativity as a media access tool, there is a possibility that negativity gets too
widespread. As a result, journalists need to draw a line about which type of negativity gains
priority over which. This is where incivility comes in. Incivility is defined as the breakage of
conventional social norms of communication (Mutz 2015; Walter 2021). While negativity usage
represents criticisms among politicians on their policies and traits (Geer 2006), incivility goes
beyond by including additional elements such as name-calling or mocking (Sobieraj and Berry
2011). Consequently, if negativity becomes uncivil, it does not only align with the news value of
conflict, but covers other values such as entertainment, where certain forms of incivility, such as
name-calling, may be perceived as humorous (Verhulsdonk et al. 2021). This makes uncivil
negativity extremely newsworthy (Muddiman 2018; Mutz 2015; Skytte 2019) provoking stronger
responses among the public in comparison to civil negative communication (e.g. Hopmann et al.
2018; Reiter and Matthes 2021; Walter and Ridout 2021).
Following this theoretical outline, I propose three main hypotheses (see Figure 1). Firstly,
I expect that negativity usage by politicians pays off in terms of gaining media access due to the
negativity bias. In other words, negative politicians will have a higher probability of appearing in
the news (H1). In addition, among negative politicians, I expect those that use incivility in the
12
process to be particularly successful in gaining media access (H2). Lastly, I argue that this media
logic has an impact on politicians through the mediatization of politics. Because politicians
experience greater media coverage when they are negative, I expect them to rely upon negativity
even more compared to politicians that never experience media access (H3).
Figure 1. Causal structures that lead to negativity usage by politicians
Methodology
I test my hypotheses using the case of Belgium, which is relevant due to the several characteristics
that make the confirmation of hypotheses less likely. Firstly, the politics of Belgium can be
described as a consensus democracy with a multi-party system where politicians need to
cooperate to pass legislation or form a government (see Lijphart 2012). As a result, negativity
usage in such countries tends to be low due to the importance of forming coalitions, unlike
majoritarian two-party systems where there is minimal cross-party cooperation (Elmelund-
Præstekær 2010). In addition, the boomerang effect of negativity is costlier in multi-party systems
such as Belgium’s. Voters in multi-party systems can easily change their vote choice for an
13
ideological proximate party, something that is less likely in two-party systems (Walter 2014;
Walter et al. 2014). Therefore, Belgian politicians may be more unlikely to use negativity as a
media access tool as it may diminish party cooperation or make voters side with the competition.
Moreover, given that Belgium has a federal political system with decentralised power,
politicians do not need to fight for media attention as much as politicians in other countries that
are more centralised. As was shown by Vos and Van Aelst (2018), politicians in Belgium can expect
to enjoy considerable media access regardless of whether they are prime ministers, ministers,
party leaders, or ordinary members of parliament (MPs). This is different from news in centralised
majoritarian systems such as the US or the UK, where the role of president or prime minister
makes up for the majority of political media coverage. Therefore, due to the decentralised
political system, individual politicians in Belgium do not have as strong internal pressure to attract
media attention through negativity as they already enjoy considerable media coverage. The
absence of individual media-seeking behaviour is further reinforced by Belgium's partocratic
system, where parties, rather than individual politicians, play a central role in the political
landscape.
Lastly, regarding media, Belgium has traditionally been classified into the Democratic
Corporatist media model which is characterised by strong journalism professionalisation with less
commercial pressure (Hallin and Mancini 2004). This limits commercial media logic that
contributes to more sensationalism in the news (Arbaoui et al. 2020). However, recent studies
suggest that Belgium might exhibit elements of other media models where journalist
professionalisation levels are lower (Humprecht et al. 2022). Nevertheless, even in these cases,
Belgium still scores high in state support for free media, a factor negatively correlated with
14
political parallelism (Humprecht et al. 2022). Therefore, regardless of how we categorise
Belgium's media model, there is an indication that the country's media offers a less fertile ground
for identifying negativity biases in the news, especially when compared to countries with low
state support and fully liberalised media landscape.
While working with the Belgian case offers several advantages, there is one notable
disadvantage to consider. Given that politicians’ negative usage to seek media attention is
uncommon, this rarity implies that when politicians do choose to go negativity, this action is more
likely to attract significant media coverage. Consequently, Belgium becomes a more viable case
for confirming H1 and H2. Nevertheless, recent studies (Poljak 2023a) reveal that the utilisation
of negativity in Belgian politics remains consistently low through time. This implies that journalists
in Belgium generally encounter consistent levels of negativity from politicians, eliminating
concerns about a sudden surge in negativity that would increase the probability of identifying
negativity bias in news.
Data
To test my hypotheses, I analyse individual politicians’ negativity usage in Belgium’s federal
parliament (De Kamer) during question time
1
(QT) and compare it with politicians’ media access
1
Question time in Belgium takes place every week for approximately two hours during which minority and majority
MPs pose questions to the prime minister, deputy prime ministers, ministers, and other cabinet members (i.e.
secretaries). Questions are asked in thematic blocks (e.g. on health), and designated cabinet members provide
answers to questions, after which all MPs that posed questions in this block may refute answers they received. The
share of questions is distributed equally to all parties, regardless of their parliamentary size.
15
in prime-time TV news in Flanders – one of the two major regions in the country. The choice of
QT is driven by its consistent occurrence in day-to-day politics, attracting high media coverage
and capturing citizens' attention on a weekly basis (Osnabrügge et al. 2021). This makes it a
suitable platform for testing politicians' use of negativity as a media-access tool on a day-to-day
basis.
I initially use the QuestionTimeSpeech dataset (Poljak and Mertens 2022), where the units
of observation represent each speech contribution during QT debates in Belgium. The strength of
this dataset is that it goes beyond parliamentary questions which are predominantly studied in
the literature (see Borghetto and Chaqués-Bonafont 2019), and also contains answers,
interruptions, and points of order. As such, besides ordinary members of parliament (MPs), this
allows us to study cabinet members as well (Prime Ministers and Ministers).
To code the usage of negativity, I randomly sampled QTs by taking one QT per month
between January 2010 and December 2020. In total, I sampled 103 QTs (30.4% of all QTs between
2010 and 2020) during which 6.634 speeches were made. Four coders underwent six weeks of
training to code negativity in the speeches, achieving Krippendorff's alpha scores exceeding .80
for all relevant variables. Negativity was operationalised as attacks in which (i) criticism is explicitly
attributed to (ii) a political actor (following Geer 2006). Coders coded negativity directed at any
formal political actor such as individuals (e.g. PM), groups of individuals (e.g. ministers), individual
parties (e.g. Greens), or a group of parties (e.g. coalition government). Negativity directed toward
informal political actors such as interest groups, the army, or foreign political actors was not
coded. Among speeches containing negativity, coders were also tested for coding incivility, which
16
was broadly defined as any form of impolite speech (Mutz 2015), including name-calling, mocking,
or sarcasm
2
.
However, it's important to acknowledge that the focus on negativity directed at formal
political actors and reliance on textual data means I overlook intricate details that can exist in
communication, such as intolerant speech directed at parts of society (Rossini 2022) or nonverbal
elements like eye-rolling (Mutz 2015). While my scope is somewhat limited, it provides a robust
and conservative estimation of the effects between politicians' usage of negativity and the media.
Overall 2.168 speeches (32.7%) contained negativity and among these negative speeches,
559 (25.8%) contained incivility. Non-negative speeches had an average of 212 words (SD = 180
words), while negative speeches averaged 246 words (SD = 123 words). Examples of negativity
and incivility usage are available in Table 1.
Table 1. Examples of coding negativity and incivility in speeches
Neg.
Incivility
Politician
QT
Speech
No (0)
No (0)
Rachid Madrane
(PS)
7.6.2012
Mr. Prime Minister, we welcome the rapid and firm position of your
government with regard to strengthening the fight against radicalism (…)
Sarah Smeyers
(N-VA)
2.4.2010
Madam Minister, I really hope from the bottom of my heart that there will be
no relaxation in the area of the guidance of the unemployed by the RVA.
2
For example, name-calling encompasses instances where politicians are referred to using terms other than their
name or official title, like calling a minister a Smurf. Mocking involves ridiculing politicians, e.g. when a minister is
labelled the Grinch Who Stole Christmas. Sarcasm is coded when politicians make ostensibly positive statements that
are undercut by criticism, as in saying, You are a Champion, followed by the statement, Champion in lying. As a result,
we coded instances of mockery or sarcasm only when they were clearly and obviously present, disregarding any types
that could be considered ambiguous or fall under a grey area.
17
Yes (1)
No (0)
Charles Michel
(MR)
8.2.2018
Madam MP the way in which your question is posed shows an imperfect or
incomplete knowledge of the way the government deals with this issue.
Monica De
Coninck (sp.a)
26.4.2018
However, I do not accept that so many children disappear in this country. I
think there should be a priority action plan, not just from you, but from the
entire government.
Yes (1)
Marco Van Hees
(PVDA)
2.7.2015
Minister, I note that in addition to playing the role of the Smurf with glasses,
you also play the role of the happy Smurf – “it will be better tomorrow”.
Tanguy Veys
(VB)
6.2.2014
You wash your hands in innocence, then no longer like a Walloon Houdini, but
like Pontius Pilate.
Note: More examples are available in Appendix B.
The testing of the hypotheses was conducted by transforming the coded data so that each
politician that spoke during a particular QT constitutes a unit of observation. In total, the final
dataset contains 2.829 observations of politicians participating in QT with 367 individual
politicians re-appearing across 103 QTs. Speakers that moderate QTs are omitted as they go
negative to get politicians back in line when they break the rules of procedures. As such, negativity
used by speakers is not suited for testing the theoretical framework of strategic political
behaviour driven by the media. Furthermore, I omitted independent politicians without party
affiliation as it is not possible to control their ideological positions, which may impact negativity
usage (e.g. Maier and Nai 2021).
3
Negativity is the first binary variable that indicates whether a particular politician during
a particular QT engaged in negativity (0=No; 1=Yes). This is followed by a second binary variable
incivility that indicates whether incivility was used in the process (0=No; 1=Yes). For example, if a
politician A on a specific QT (1 observation in data) attacked another political actor for a policy
3
Despite omitting participation of speakers (N=103) and independent MPs (N=65) in QTs, sensitivity analyses
including these groups of politicians do not show different results from the ones reported in the main text.
18
they advocated, a value in the negativity variable is 1 indicating that politician A engaged in
negativity. In addition, if politician A also mocked the other actor in the process, a value of 1 is
attributed to the variable incivility. On average, 45.4% of politicians per QT engage in negativity
and the average incivility usage per QT (among politicians that engage in negativity) is 31%. These
numbers are consistent through time (see yearly trends in Appendix B). As such, the dataset
includes a variety of politicians that go negative but also many politicians that remain neutral in
a debate by not criticising anyone explicitly. Given that politicians may attack more than one actor
on a particular QT, I also generate additional count variables indicating the total amount of
negativity and incivility used, which will be used in the sensitivity analyses (Appendix C).
Media access is the third binary variable and shows whether a politician following a QT
session (in which it participated) also got media access (0=No; 1=Yes). To generate this variable, I
used the Electronic News Archive (ENA) dataset from Belgium, which collects appearances of
politicians in prime-time 7 pm Flemish news on the main public (VRT) and private (VTM)
broadcasts in Flanders
4
(nieuwsarchief.be). This variable is placed future in time, more
specifically, +7 hours following a QT in the parliament. On average, 14.4% of politicians that
participate in QT get a chance to appear in the prime-time news following QT. This shows that
media access does not come hand in hand with simply participating in QT. Additional binary
variables for only public or private media access are also generated, together with the overall
4
The market is evenly split between the two while serving almost 80% of the 6 million Flemish audiences in Belgium
(De Swert and Hooghe 2010). According to the Reuters Institute Digital News Report from 2022, over 70% of the
Flemish audience trusts both prime-time TV news while 60% indicate TV as a source of news (Reuters Institute 2022).
There is no national television in Belgium.
19
count of media appearances following QT, all of which will be used in sensitivity analyses
(Appendix C).
Cumulative media access is the fourth variable. This variable is a cumulative variable and
indicates for each observation the number of times the politician gained media access due to
negativity in previous observations. If a politician was negative three times in the past (Negativity
= 1), and two times this negativity resulted in media coverage (Media Access = 1), this politician
receives a cumulative media access of 2 at that point in time. There are 748 observations of
politicians that have non-zero cumulative media access.
5
Additionally, a binary variable is
generated separating politicians into a control group (no media access following negativity ever)
and a treatment group (did experience media access following negativity at least once). This
variable will be used in the sensitivity analyses (Appendix C).
Analysis
Due to the dichotomous nature of the dependent variables (negativity and media access), the
testing of the hypotheses was conducted by running logistic regressions. Three main models are
run, one for each hypothesis. To test H1, i.e. media granting more access to negative politicians,
media access is the main dependent variable placed at t (7 pm TV news), whereas negativity is
5
51.5% of these politicians have a cumulative media access of 1, 20% have a cumulative media access of 2, 8.6% have
a cumulative media access of 3, 6.5% have a cumulative media access of 4, 5.8% have a cumulative media access of
5, and 4.7% have a cumulative media access of 6. All observations above 6 make up less than 1% of observations with
a maximum value of 11.
20
used as the main independent variable and is lagged at t-1 (seven hours
6
before TV news). To test
H2, i.e. politicians that use uncivil negativity are more likely to gain media access compared to
politicians that use civil negativity, I repeat the previous model but instead of negativity, I use
incivility in negativity as the main independent variable. Therefore, in this particular model, I only
work with observations of politicians per QT that displayed negative behaviour during QT
(N=1,284). Lastly, to test H3, i.e. politicians who get media access due to negativity are more likely
to go negative again, I take negativity as the dependent variable and place it at t (QT), while
cumulative media access is placed at t-1 (all observations before the observed QT). In this final
model, I remove all politicians that appear only once or that were never negative in my data
leading to the final N of 2,537. This makes the model more valid as it evaluates observations of
politicians that (no not) accumulate media access following negativity.
All models were run through a multi-level modelling structure in which individual
politicians (N=367) and QTs (N=103) are inserted as a random intercept but are crossed with each
other (see Figure 2). This modelling structure is important due to the panel nature of data in which
politicians re-appear as observations (see more in Chung and Beretvas 2012). As such,
observations are clustered under politicians that re-appear through QT sessions, safeguarding
that no individual or QT is skewing the results (e.g. a specific politician who generally goes
negative during QT and gets into the news often). In addition to the modelling structure, each
regression includes several control variables that may impact media access of MPs and Cabinet
Members (Yildirim et al. 2023) and negativity/incivility usage (e.g. Goovaerts and Turkenburg
6
The choice of the 7-hour lag is the result of journalists actively following the debate in real-time (e.g., the special TV
program Villa Politica during QT), making it plausible that politicians from QT appear in the news on the same day.
21
2022). This includes the status of the politicians (Opposition MP; Majority MP; Minister; Deputy
PM; PM), their gender (Man; Woman), and their parties’ ideological extremity (distance from the
ideological centre based on the Chapel Hill Expert Data; Jolly et al. 2022). Furthermore, due to the
importance of the electoral cycle (e.g. Poljak, 2023a), I also control election proximity (how many
months have passed since the last election) and also for regional differences in Belgium (Flanders;
Wallonia). I also include public approval of politicians’ parties in my models, but due to the data
scarcity on approval in Belgium, which removes almost 40% of data observations, this variable is
only included in the sensitivity analyses (Appendix C).
Figure 2. Visual illustration of the multi-level model crossing politicians with QTs in which they
participate
Results
The results of the multi-level logistic regressions are reported in Table 2. These results show a
significant negativity (and to a lesser extent incivility) bias in the news when politicians are
covered confirming H1 and H2. More specifically, Model 1 shows a positive and significant
coefficient for the negativity variable: politicians that go negative in the parliament at noon are
more likely to appear in the prime-time TV news later that day at 7 pm when compared to
22
politicians that do not engage in negativity. In addition, Model 2 provides support that among
those that do go negative, politicians that use incivility in the process are significantly more likely
to appear in the news when compared to those that used negativity that was civil. Regarding H3,
the results also confirm the mediatization of (negative) politicians. A positive and significant
coefficient for the cumulative media access variable shows that the more media access politicians
accumulate due to their negative behaviour, the more likely they become to use negativity again.
In addition to the main predictors, the results also highlight that cabinet members enjoy
significantly greater media access (see also Yildirim et al. 2023) and men politicians gain more
media access following QT when compared to women politicians (Thesen and Yildirim 2023).
Table 2. Testing the probability of gaining media access (Model 1;2) and using negativity (Model
3)
Model 1
Model 2
Model 3
DV: Media-access (1=Yes) (t)
DV: Negativity (1=Yes) (t)
Coef. (SE)
Coef. (SE)
Independent variables
Negativity (ref. no) (t-1)
.787 (.174) ***
-
-
Incivility in negativity (ref. no) (t-1)
-
.629 (.236) **
-
Cumulative media access (t-1)
-
-
.111 (.044) *
Control variables
Electoral cycle
.004 (.006)
.008 (.009)
.006 (.004)
Opposition MP (ref.)
Majority MP
.048 (.248)
.291 (.316)
-2.005 (.134) ***
Minister
2.090 (.293) ***
2.253 (.481) ***
-3.067 (.203) ***
Deputy PM
2.661 (.344) ***
2.029 (.586) **
-3.290 (.266) ***
PM
5.998 (.593) ***
6.639 (1.028) ***
-2.285 (.358) ***
Ideology
-.744 (1.029)
-1.930 (1.292)
1.737 (.578) **
Man (ref.)
Woman
-.482 (.216) *
-.569 (.295) †
-.011 (.122)
Flanders (ref.)
Wallonia
-1.907 (.241) ***
-2.271 (.364) ***
-.247 (.122) *
Constant
-2.817 (.395) ***
-2.191 (.513) ***
.866 (.224) ***
23
Variance (QTs)
.582 (.196)
.568 (.284)
.176 (.063)
Variance (Politicians)
.337 (.118)
.798 (.337)
.119 (.064)
N (observations)
2.829
1.284
2.537
N (QTs)
103
103
103
N (individual politicians)
367
269
234
N (min. politicians per QT)
13
5
13
N (max. politicians QT)
37
22
35
AIC (empty model)
1.748 (0=1.930)
789 (0=887)
2.659 (0=3.031)
Note: †p<0.1 *p<0.05; **p<0.01; ***p<0.00
To get a visualisation of these findings, post-estimated predicted probabilities of
regression analyses are provided in Figure 3. Regarding negativity bias in the news, shown in the
top-left side, the average probability of media access at 7 pm TV news is 60.7% more likely for
politicians that were negative during QT compared to those that were not (9.5 to 15.4). Zooming
in on those that were negative, at the top-right side of Figure 3, the average probability for media-
access increases by 53.3% for politicians that decided to use incivility (8.1 to 12.4). Finally, at the
bottom of Figure 3, we can also see that for each media access following negativity, the average
probability that a politician goes negative again increases. For example, politicians that w ere in
the news twice after they were negative have a 4.5% higher probability of going negative again
compared to those that were in the news following negativity only once (45.5 to 47.6).
24
Figure 3. Predicted probability of gaining media access following negativity in QT (top-left), using
incivility in negativity (top-right), and going negative again following media access after negativity
(bottom)
Note: Vertical lines indicate 90% confidence intervals holding other variables at their mean.
Further Tests of Causality
Before discussing and concluding the results, it is necessary to address several issues related to
determining causality. The first issue of causality arises from politicians using negativity on issues
that the media seeks to cover, indicating that it is issues, rather than negativity, that drive
journalists' bias towards negative politicians. To explore this, speeches were automatically issue-
25
coded using the Comparative Agendas Project dictionary. This dictionary maps references to 21
major issues, such as the economy, immigration, social welfare, or defence (Albaugh et al. 2013).
The resulting dataset consists of observations for each issue (N=21) within a given QT (N=103),
indicating the share of attention devoted to each issue (N=2,163). By testing the relationship
between issue attention in the two groups of speeches (those that were negative vs. those that
were not), a positive and significant correlation was found (Pearson's r = .5419; p = .000).
Therefore, although not perfectly correlated, this finding supports the notion of a negative bias
among journalists, as they may cover politicians who are not negative in QT but address similar
issues as those who adopt negativity.
The second causality issue relates to the possibility of consistently negative politicians,
creating endogeneity in Model 3 (Table 2). This means that media reinforcing negativity among
politicians may not be driven by the media itself, but by politicians with a consistent tendency
toward negativity. To investigate this, I focused on 234 politicians who displayed negativity at
least once in the dataset. They were split into two groups: a control group with no media access
after negativity, and a treatment group with media access following negativity (using the sa me
model as in Table 2, i.e., Model 2). The results (full output in Appendix C3) revealed a significant
difference between the two groups. The probability of politicians adopting negativity increased
by 17.8% (from .45 to .53; see Figure 4) when comparing the control group (no media access) to
the treatment group (media access) following negativity.
26
Figure 4 Predicted probability of going negative if a politician has no experience with media access
following negativity (control) vs. politicians that do have experience following negativity
(treatment)
Note: Vertical lines indicate 90% confidence intervals holding other variables at their mean.
Another endogeneity concern is the possibility that politicians respond to media scrutiny
by adopting a negative tone in QTs to protect their reputation. To investigate this, I examined
politicians' cumulative media exposure following QTs when they were not negative. The analysis
revealed no significant relationship, indicating that being in the news after QTs does not result in
increased negativity in subsequent QTs (Appendix C4). This finding further supports H3 and
suggests that being in the news doesn't inherently stimulate politicians to be negative; rather, it's
their own supply of negativity, if picked up by journalists, that stimulates subsequent negative
rhetoric. While the issue of endogeneity does remain, these findings do provide additional
evidence that the media may act as a reinforcing role for politicians' negativity usage.
Sensitivity analyses
27
Lastly, to further ensure the validity of the main findings, several sensitivity analyses were
conducted (Appendix C1-C7). These analyses examined differences between public and private
TV broadcasts, tested the hypotheses using count dependent variables, including public approval,
omitting 2020 due to the round of the flag effect during the COVID pandemic (Louwerse et al.
2021), and finally, running fixed-effects models. These analyses consistently aligned with the
theory, but the public broadcaster in Flanders showed a greater susceptibility to negativity bias
compared to private broadcasts. This is likely due to the heightened responsibility of public
broadcasts to report on parliamentary activities (see more in Appendix C), in contrast to private
TV news.
Discussion and Conclusion
When politicians decide to go negative, voters will not immediately be inclined to vote for them.
However, politicians persistently rely on negativity. In this paper, I offered one possible
explanation for this counterintuitive behaviour: politicians use negativity as a media access tool.
Relying on the mediatization literature, I argue that politicians adopt media logic that is driven by
a negativity bias. If politicians aim to be successful in obtaining news coverage (e.g., to advance
their personal careers and their policy goals), they need to provide the media with negativity.
Using longitudinal data on the negativity usage of Belgian politicians in the country’s federal
parliament, and data on the media access in prime-time TV news, I show support for these
hypotheses. The results indicate that the media reports more on negative politicians. The more
the media prioritise the coverage of negative politicians, the more they stimulate further negative
political behaviour.
28
While these findings may initially raise concerns, there are several encouraging aspects to
consider in the results presented in this paper. Firstly, it is noteworthy that a significant number
of politicians maintain a neutral stance and refrain from engaging in negativity during
parliamentary debates in day-to-day politics (Appendix B). Moreover, even among those who do
employ negativity, the majority do so without resorting to incivility. This indicates that the
negative communication utilised by politicians predominantly adhere to social norms. Therefore,
it is important to recognise that while politicians adapt to media logic by supplying negativity,
they do so in a manner that aligns with normative expectations and can actually benefit voters
(Geer 2006; Van Elsas and Fiselier 2023). In addition, this study also indicates that negativity
among politicians is not as widespread as we may assume and is not increasing (see also
Goovaerts and Turkenburg 2022).
Furthermore, this study provides support for the existence of a negativity bias in political
news coverage, which plays a crucial role in ensuring the quality of democracy (Soroka 2014).
Specifically, when the media delivers politicians' criticisms, it empowers voters to make well-
informed decisions during elections, potentially increasing satisfaction with democracy (Ridge
2022; Tuttnauer 2022; Van Elsas and Fiselier 2023) and their turnout at elections (Schuck et al.
2016). Consequently, the media and journalists prioritising politicians who engage in negativity,
even if it doesn't always result in higher audience engagement, can be seen as a positive
indication of effectively functioning democracy.
One issue, however, is that incivility also gathers attention, howbeit, to a lesser extent
than general negativity usage (Appendix C). As such, journalistic routines within newsrooms, such
as the pressure to publish stories quickly (e.g. Bartholomé et al. 2015), may amplify incivility, even
29
when it's rarely present in practice. For instance, because incivility is easily distinguishable in a
parliamentary debate from civil discourse, journalist’s time constraints may lead to the rapid
dissemination of such rhetoric, unlike more beneficial general negativity.
Addressing these practices and allocating additional resources to modify existing
newsroom routines could help reduce the urgency to publish the most attention-grabbing yet
potentially harmful political content in the news. This is crucial because uncivil rhetoric has been
demonstrated to diminish the quality of political debates (Marien et al. 2020), leading to
decreased levels of political trust (Van't Riet and Van Stekelenburg 2022). Consequently, when
citizens are exposed to political incivility in the news, it may reinforce their resentment of politics
(e.g. Fridkin and Kenney 2011; Gervais 2017; Hopmann et al. 2018; Reiter and Matthes 2021;
Walter and Ridout 2021).
Yet, one should be cautious in generalising these findings, as they only apply to the multi-
party and non-liberal media model of Belgium. Still, there are reasons to expect that the results
from this paper do apply to other systems, like the US. For instance, the two-party system tends
to be more negative (e.g. Elmelund-Præstekær 2010), and the commercial media landscape
enhances the negativity bias (e.g. Sacerdote et al. 2020). Therefore, this paper aims to stimulate
a broader debate regarding the link between politicians' negativity usage and the influence of
media, driven by audience engagement, in reinforcing such behaviour. Acquiring a deeper
understanding of this subject is crucial as it enables us to gain insights into the quality of
democracies, which necessitates the presence of negative political communication (Geer 2006).
Additionally, this knowledge helps us comprehend the challenges faced by contemporary
30
democracies, such as declining political trust and escalating affective polarization. By delving into
these aspects, we can foster greater awareness and facilitate potential solutions for these issues.
References
Albaugh, Quinn, Julie Sevenans, Stuart Soroka, and Peter John Loewen (2013). The automated
coding of policy agendas: A dictionary-based approach. Presented at the 6th Annual
Comparative Agendas Conference. Antwerp, Belgium.
Arbaoui, Bouchra, Knut De Swert, and Wouter Van der Brug (2020). Sensationalism in news
coverage: A comparative study in 14 television systems. Communication Research, 47(2),
299-320.
Bartholomé, Guus, Sophie Lecheler, and Claes de Vreese (2015). Manufacturing conflict? How
journalists intervene in the conflict frame building process. The International Journal of
Press/Politics, 20(4), 438-457.
Benoit, William L. (1999). Seeing spots: A functional analysis of presidential television
advertisements, 1952-1996. Greenwood Publishing Group.
Benoit, William L., Kevin A. Stein, and Glenn J. Hansen (2004). Newspaper coverage of presidential
debates. Argumentation and Advocacy, 41(1), 17-27.
Borghetto, Enrico, and Laura Chaqués-Bonafont (2019). Parliamentary questions. In
Baumgartner, F., Breunig, C. and Grossman, E. (eds) Comparative Policy Agendas, Oxford,
Oxford University Press, pp. 282–405.
31
Chung, Hyewon, and S. Natasha Beretvas (2012). The impact of ignoring multiple membership
data structures in multilevel models. British Journal of Mathematical and Statistical
Psychology, 65(2), 185-200.
Conway, Bethany A., Kate Kenski, and Di Wang (2015). The rise of Twitter in the political
campaign. Journal of computer-mediated communication, 20(4), 363-380.
Damore, David F. (2002). Candidate strategy and the decision to go negative. Political Research
Quarterly, 55(3), 669-685.
Dassonneville, Ruth (2023). Voters Under Pressure: Group-Based Cross-Pressure and Electoral
Volatility. Oxford University Press.
De Swert, Knut, and Marc Hooghe (2010). When do women get a voice? Explaining the presence
of female news sources in Belgian news broadcasts (2003—5). European Journal of
Communication, 25(1), 69-84.
Djerf-Pierre, Monika, and Adam Shehata (2017). Still an agenda setter. Journal of Communication,
67(5), 733-757.
Druckman, James N., et al. (2019). How incivility on partisan media (de) polarizes the electorate.
The Journal of Politics, 81(1), 291-295.
Dubois, Elizabeth, and Grant Blank (2018). The echo chamber is overstated. Information,
communication & society, 21(5), 729-745.
Elmelund-Præstekær, Christian (2010). Beyond American negativity: Toward a general
understanding of the determinants of negative campaigning. European Political Science
Review, 2(1), 137-156.
32
Elmelund-Præstekær, Christian, David Nicolas Hopmann, and Asbjørn Sonne Nørgaard (2011).
Does mediatization change MP-media interaction and MP attitudes toward the media?
Evidence from a longitudinal study of Danish MPs. The International Journal of
Press/Politics, 16(3), 382-403.
Fournier, Patrick, Stuart Soroka, and Lilach Nir (2020). Negativity biases and political ideology: A
comparative test across 17 countries. American Political Science Review, 114(3), 775-791.
Fridkin, Kim L., and Patrick Kenney (2004). Do negative messages work? The impact of negativity
on citizens’ evaluations of candidates. American Politics Research, 32(5), 570-605.
Fridkin, Kim L., and Patrick Kenney (2011). Variability in citizens’ reactions to different types of
negative campaigns. American Journal of Political Science, 55(2), 307-325.
Frimer, Jeremy A., et al. (2023). Incivility is rising among American politicians on Twitter. Social
Psychological and Personality Science, 14(2), 259-269.
Garzia, Diego, Frederico Ferreira da Silva, and Andrea De Angelis (2022). Partisan dealignment
and the personalisation of politics in West European parliamentary democracies, 1961–
2018. West European Politics, 45(2), 311-334.
Geer, John G. (2006). In Defense of Negativity: Attack Ads in Presidential Campaigns. Chicago and
London: University of Chicago Press.
Geer, John G. (2012). The news media and the rise of negativity in presidential campaigns. PS:
Political Science & Politics, 45(3), 422-427.
Gervais, Bryan T (2017). More than mimicry? The role of anger in uncivil reactions to elite political
incivility. International Journal of Public Opinion Research, 29(3), 384-405.
33
Goovaerts, Ine, and Sofie Marien (2020). Uncivil communication and simplistic argumentation:
Decreasing political trust, increasing persuasive power?. Political Communication, 37(6),
768-788.
Goovaerts, Ine, and Emma Turkenburg (2022). How Contextual Features Shape Incivility Over
Time: An Analysis of the Evolution and Determinants of Political Incivility in Televised
Election Debates (1985–2019). Communication Research.
Hallin, Daniel C., and Paolo Mancini (2004). Comparing media systems: Three models of media
and politics. Cambridge University Press.
Harcup, Tony, and Deirdre O’Neill. (2017). What is news? News values revisited (again).
Journalism studies, 18(12), 1470-1488.
Harrington Jr, Joseph E., and Gregory D. Hess (1996). A Spatial Theory of Positive and Negative
Campaigning. Games and Economic Behavior, 17(2), 209–29.
Haselmayer, Martin, Thomas M. Meyer, and Markus Wagner (2019). Fighting for attention: Media
coverage of negative campaign messages. Party Politics, 25(3), 412-423.
Haynes, Audrey A., and Staci L. Rhine (1998). Attack politics in presidential nomination campaigns:
An examination of the frequency and determinants of intermediated negative messages
against opponents. Political Research Quarterly, 51(3), 691-721.
Humprecht, Edda, et al. (2022). Media systems in the digital age: An empirical comparison of 30
countries. Journal of Communication, 72(2), 145-164.
Hopmann, David Nicolas, Rens Vliegenthart, and Jürgen Maier (2018). The effects of tone, focus,
and incivility in election debates. Journal of Elections, Public Opinion and Parties, 28(3),
283-306.
34
Iyengar, Shanto, Gaurav Sood, and Yphtach Lelkes (2012). Affect, not ideology social identity
perspective on polarization. Public opinion quarterly, 76(3), 405-431.
Jennings, Will, and Christopher Wlezien (2018). Election polling errors across time and space.
Nature Human Behaviour, 2(4), 276-283.
Jolly, Seth et al. (2022). Chapel Hill expert survey trend file, 1999–2019. Electoral studies, 75.
Jürgens, Pascal, Andreas Jungherr, and Harald Schoen. (2011). Small worlds with a difference:
New gatekeepers and the filtering of political information on Twitter. In Proceedings of
the 3rd international web science conference (pp. 1-5).
Ketelaars, Pauline (2019). Position, Preference and Personality: A Microlevel Explanation of
Negativity in Day‐To‐Day Politics. Political Psychology, 40(5), 1019-1038.
Knobloch-Westerwick, Silvia, Cornelia Mothes, and Nick Polavin (2020). Confirmation bias,
ingroup bias, and negativity bias in selective exposure to political information.
Communication Research, 47(1), 104-124.
Lamot, Kenza, and Peter Van Aelst (2020). Beaten by Chartbeat? An experimental study on the
effect of real-time audience analytics on journalists’ news judgment. Journalism Studies,
21(4), 477-493.
Lau, Richard R., Lee Sigelman, and Ivy Brown Rovner (2007). The effects of negative political
campaigns: A meta-analytic reassessment. The Journal of Politics, 69(4), 1176-1209.
Lengauer, Günther, Frank Esser, and Rosa Berganza (2012). Negativity in political news: A review
of concepts, operationalizations and key findings. Journalism, 13(2), 179-202.
Lijphart, Arend (2012). Patterns of democracy. Second Edition. Yale University Press.
35
Lipsitz, Keena, and John G. Geer (2017). Rethinking the concept of negativity: An empirical
approach. Political Research Quarterly, 70(3), 577-589.
Louwerse, Tom, and Simon Otjes (2016). Personalised parliamentary behaviour without electoral
incentives: the case of the Netherlands. West European Politics, 39(4), 778-799.
Louwerse, Tom, et al. (2021). Opposition in times of crisis: COVID-19 in parliamentary debates.
West European Politics, 44(5-6), 1025-1051.
Maier, Jürgen, and Carolin Jansen (2017). When do candidates attack in election campaigns?
Exploring the determinants of negative candidate messages in German televised debates.
Party Politics, 23(5), 549-559.
Maier, Jürgen, and Alessandro Nai (2020). Roaring candidates in the spotlight: Campaign
negativity, emotions, and media coverage in 107 national elections. The International
Journal of Press/Politics, 25(4), 576-606.
Maier, Jürgen, and Alessandro Nai (2021). Mapping the drivers of negative campaigning: Insights
from a candidate survey. International Political Science Review.
Maier, Jürgen, Sebastian Stier, and Corinna Oschatz (2022). Are candidates rational when it comes
to negative campaigning? Empirical evidence from three German candidate surveys. Party
Politics.
Maisel, L. Sandy (2012). The negative consequences of uncivil political discourse. PS: Political
Science & Politics, 45(3), 405-411.
Marien, Sofie, Ine Goovaerts, and Stephen Elstub (2020). Deliberative qualities in televised
election debates: the influence of the electoral system and populism. West European
Politics, 43(6), 1262-1284.
36
Muddiman, Ashley (2018). Attributions of incivility in presidential campaign news. In Warner, B.
et al. (eds.) An Unprecedented Election (pp. 115–133). Praeger.
Muddiman, Ashley, Jamie Pond-Cobb, and Jamie E. Matson. (2020). Negativity bias or backlash:
Interaction with civil and uncivil online political news content. Communication Research,
47(6), 815-837.
Mueller, Samuel David, and Marius Saeltzer (2022). Twitter made me do it! Twitter's tonal
platform incentive and its effect on online campaigning. Information, Communication &
Society, 25(9), 1247-1272.
Mutz, Diana C. (2015). In Your Face Politics. Princeton University Press.
Nai, Alessandro (2013). What really matters is which camp goes dirty: Differential effects of
negative campaigning on turnout during Swiss federal ballots. European Journal of Political
Research, 52(1), 44-70.
Nai, Alessandro, and Annemarie Walter (2015) (eds). New Perspectives on Negative
Campaigning. Colchester: ECPR Press.
Nai, Alessandro, and Pascal Sciarini (2018). Why “going negative?” Strategic and situational
determinants of personal attacks in Swiss direct democratic votes. Journal of Political
Marketing, 17(4), 382-417.
Nai, Alessandro, and Jürgen Maier (2021). Is negative campaigning a matter of taste? Political
attacks, incivility, and the moderating role of individual differences. American Politics
Research, 49(3), 269-281.
37
Nai, Alessandro, Anke Tresch, and Jürgen Maier (2022). Hardwired to attack. Candidates’
personality traits and negative campaigning in three European countries. Acta Politica,
57(4), 772-797.
Niven, David (2001). Bias in the news: Partisanship and negativity in media coverage of presidents
George Bush and Bill Clinton. Harvard International Journal of Press/Politics, 6(3), 31-46.
Oleskog Tryggvason, Per (2020). How Mediated Opinion Polls Influence Political Parties: Revisiting
the Arena Framework. International Journal of Public Opinion Research, 32:2, 243–65.
Osnabrügge, Moritz, Sara B. Hobolt, and Toni Rodon (2021). Playing to the gallery: Emotive
rhetoric in parliaments. American Political Science Review, 115(3), 885-899.
Peeters, Jeroen. (2022). What's the issue? Studying supply and demand side of online issue
strategies of politicians. Universiteit Antwerpen.
Poljak, Željko (2023a). Parties’ attack behaviour in parliaments: Who attacks whom and when.
European Journal of Political Research, 62(3), 903-923.
Poljak, Željko (2023b). The impact of public approval on the use of negativity throughout the
electoral cycle. West European Politics, 46(5), 947–970.
Poljak, Željko, and Tim Mertens (2022). QuestionTimeSpeech v.1. Harvard Dataverse.
Reiter, Franz, and Jörg Matthes (2021). “The Good, the Bad, and the Ugly”: A Panel Study on the
Reciprocal Effects of Negative, Dirty, and Positive Campaigning on Political Distrust. Mass
Communication and Society, 1-24.
Reuters Institute (2022). Digital News Report 2022. Available at:
https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2022-06/Digital_News-
Report_2022.pdf
38
Robertson, Claire E., et al. (2023). Negativity drives online news consumption. Nature Human
Behaviour.
Ridge, Hannah M. (2022). Just like the others: Party differences, perception, and satisfaction with
democracy. Party Politics, 28(3), 419-430.
Ridout, Travis N., and Annemarie S. Walter (2015). How the news media amplify negative
messages. In A. Nai and A. Walter (eds.) New Perspectives on Negative Campaigning (pp.
267-286). ECPR Press.
Rossini, Patrícia (2022). Beyond incivility: Understanding patterns of uncivil and intolerant
discourse in online political talk. Communication Research, 49(3), 399-425.
Sacerdote, Bruce, Ranjan Sehgal, and Molly Cook (2020). Why is all COVID-19 news bad news?.
National Bureau of Economic Research.
Schuck, Andreas RT, Rens Vliegenthart, and Claes H. De Vreese. (2016). Who's afraid of conflict?
The mobilizing effect of conflict framing in campaign news. British Journal of Political
Science, 46(1), 177-194.
Sevenans, Julie, Stefaan Walgrave, and Debby Vos (2015). Political elites’ media responsiveness
and their individual political goals: A study of national politicians in Belgium. Research &
Politics, 2(3).
Shoemaker, Pamela J. (1996). Hardwired for news: Using biological and cultural evolution to
explain the surveillance function. Journal of communication, 46(3), 32-47.
Shoemaker, Pamela J., Tim P. Vos, and Stephen D. Reese (2009). Journalists as gatekeepers. In
The handbook of journalism studies (pp. 93-107). Routledge.
39
Skaperdas, Stergios, and Bernard Grofman (1995). Modeling negative campaigning. American
Political Science Review, 89(1), 49-61.
Skytte, Rasmus (2019). Why politicians seem so rude and how it affects citizens. Doctoral thesis.
Aarhus University.
Skytte, Rasmus (2022). Degrees of Disrespect: How Only Extreme and Rare Incivility Alienates the
Base. The Journal of Politics, 84(3), 1746-1759.
Sobieraj, Sarah, and Jeffrey M. Berry (2011). From incivility to outrage: Political discourse in blogs,
talk radio, and cable news. Political Communication, 28(1), 19-41.
Somer-Topcu, Zeynep, and Daniel Weitzel (2022). Negative Campaigning and Vote Choice in
Europe: How Do Different Partisan Groups React to Campaign Attacks?. Comparative
Political Studies, 55(13), 2283-2310.
Soontjens, Karolin (2021). The awareness paradox:(Why) politicians overestimate citizens’
awareness of parliamentary questions and party initiatives. Representation, 57(1), 75-94.
Soroka, Stuart (2012). The gatekeeping function: Distributions of information in media and the
real world. The Journal of Politics, 74(2), 514-528.
Soroka, Stuart (2014). Negativity in democratic politics: Causes and consequences. Cambridge
University Press.
Soroka, Stuart, and Stephen McAdams (2015). News, politics, and negativity. Political
communication, 32(1), 1-22.
Soroka, Stuart, Patrick Fournier, and Lilach Nir (2019). Cross-national evidence of a negativity bias
in psychophysiological reactions to news. Proceedings of the National Academy of Sciences,
116(38), 18888-18892.
40
Strömbäck, Jesper (2008). Four phases of mediatization: An analysis of the mediatization of
politics. The international journal of press/politics, 13(3), 228-246.
Thesen, Gunnar, and Tevfik Murat Yildirim. (2023). Electoral systems and gender inequality in
political news. American Political Science Review, 117(2), 575-590.
Thorson, Esther, et al. (2000). Negative political ads and negative citizen orientations toward
politics. Journal of Current Issues & Research in Advertising, 22(1), 13-40.
Tuttnauer, O. (2022). Satisfaction with Democracy and Government-Opposition Relations.
Working paper. Available at: https://osf.io/942ab/download (1.7.2023)
Van Aelst, Peter et al. (2008). Politicians in the news: Media or party logic? Media attention and
electoral success in the Belgian election campaign of 2003. European Journal of
Communication, 23(2), 193-210.
van Elsas, Erika, and Toine Fiselier (2023). Conflict or choice? The differential effects of elite
incivility and ideological polarization on political support. Acta Politica, 1-30.
Van Erkel, Patrick FA, and Peter Thijssen (2016). The first one wins: Distilling the primacy effect.
Electoral Studies, 44, 245-254.
Van Remoortere, Annelien, Stefaan Walgrave, and Rens Vliegenthart (2023). Mass media
occurrence as a political career maker. The International Journal of Press/Politics, 28(1),
201-218.
Van’t Riet, Jonathan, and Aart Van Stekelenburg (2022). The effects of political incivility on
political trust and political participation: A meta-analysis of experimental research. Human
Communication Research, 48(2), 203-229.
41
Verhulsdonk, Iris, Alessandro Nai, and Jeffrey A. Karp. "Are political attacks a laughing matter?
Three experiments on political humor and the effectiveness of negative campaigning."
Political Research Quarterly 75, no. 3 (2022): 720-737.
Vliegenthart, Rens, Hajo G. Boomgaarden, and Jelle W. Boumans (2011). Changes in political news
coverage: Personalization, conflict and negativity in British and Dutch newspapers. In
Brants, K. & Voltmer, K. (ed.) Political communication in postmodern democracy (pp. 92-
110). Palgrave Macmillan, London.
Vliegenthart, Rens et al. (2016). Do the media set the parliamentary agenda? A comparative study
in seven countries. European Journal of Political Research, 55(2), 283-301.
Vos, Debby (2014). Which politicians pass the news gates and why? Explaining inconsistencies in
research on news coverage of individual politicians. International journal of
communication, 8, 2438-2461.
Vos, Debby, and Peter Van Aelst (2018). Does the political system determine media visibility of
politicians? A comparative analysis of political functions in the news in sixteen countries.
Political Communication, 35(3), 371-392.
Walgrave, Stefaan (2008). Again, the almighty mass media? Political Communication, 25(4), 445-
459.
Walgrave, Stefaan, and Karolin Soontjens (2023). How politicians learn about public opinion.
Research & Politics, 10(3).
Walter, Annemarie S. (2014). Choosing the enemy: Attack behaviour in a multiparty system. Party
Politics, 20(3), 311-323.
42
Walter, Annemarie S. (2021). The study of political incivility and its challenges. In Walter, A. (ed.)
Political Incivility in the Parliamentary, Electoral and Media Arena (pp. 1–16). Routledge.
Walter, Annemarie S., and Wouter Van der Brug (2013). When the gloves come off: Inter-party
variation in negative campaigning in Dutch elections, 1981–2010. Acta Politica, 48(4), 367-
388.
Walter, Annemarie S., et al. (2014). Party competition and negative campaigning. Comparative
Political Studies, 47(4), 550-573.
Walter, Annemarie S., and Cees van der Eijk (2019). Unintended consequences of negative
campaigning: Backlash and second-preference boost effects in a multi-party context. The
British Journal of Politics and International Relations, 21(3), 612-629.
Walter, Annemarie S., and Travis Ridout (2021). Voters' emotional response to negative campaign
messages. In Walter, A. (ed.) Political Incivility in the Parliamentary, Electoral and Media
Arena (pp. 124-142). Routledge.
Wlezien, Christopher (1995). The public as thermostat: Dynamics of preferences for spending.
American Journal of Political Science, 981-1000.
Yildirim, Tevfik Murat, et al. (2023). The determinants of the media coverage of politicians:
European Journal of Political Research.
