Is it Real Life, or is it Just Fantasy?
Assessing the Processing and Persuasiveness of News Parody
Isabella van der Vlies - 11291729 University of Amsterdam Supervised by mw.dr.F.Marquart
Master’s Thesis
Graduate School of Communication Master’s Programme Communication Science
Acknowledgements
During the writing of my master’s thesis there were several people who helped and supported me. I cannot name everyone here, but there are some people I want to thank in particular. First, my supervisor Franziska Marquart. Thank you for your detailed feedback, while simultaneously giving me the freedom in conducting the study the way I wanted to. Mark Boukes, my ‘unofficial’ supervisor. Thank you for not only being a tremendous source of inspiration in the master’s programme Political Communication & Journalism, but also for steering me in the right direction throughout the writing of my thesis. I really appreciated your thoughts on the study. My colleagues from Respondenten.nl: Rik Salomé, Jeffrey Rupp and Camillus op het Veld. Thank you for giving me the opportunity to use the panel of
Respondenten.nl, which provided me with a great sample. I would like to thank my friends, and in particular Viola Dijkman. Thank you for your endless support and faith in me. Last but not least, my mom. Thanks, mom, for the numerous spelling and grammar checks throughout my whole study career: I hope you’re proud of me.
Abstract
News parody websites gain increasing popularity and are considered an important element of the current media landscape. Yet studies determining the effects of news parody websites are inadequately represented in the communication science literature. In the Netherlands, De Speld is seen as the largest and most popular satirical news website. An experimental investigation with a 3 (humour: no satire, horatian, juvenalian) x 2 (ability: low, high) between-subjects design was conducted to explore the processing and persuasiveness of
De Speld among Dutch citizens (N = 276). The effects of news parody were investigated in
relation to message resistance. As hypothesized two message-resisting strategies play a key role in the processing of news parody: 1) message discounting, and 2) source derogation. Message discounting is activated when the article is perceived as funny, and when the tone of the satirical piece is light-hearted. Moreover, the current study takes media credibility into account by including source derogation, and is directly activated when participants are
familiar with the comic intent of De Speld. Additional testing concludes that the processing of news parody is a complex interplay between different variables. Furthermore, older
participants engage in message-resisting strategies to a lesser extent than younger individuals. Ultimately the current study finds attitudinal and behavioural effects of news parody to be limited. Methodological limitations of the present study and implications for further research are discussed.
Keywords: news parody, De Speld, political satire, processing, persuasiveness,
message-resisting, source derogation, message discounting
word count: 7346 (excluding: acknowledgments, abstract, references, endnotes and appendices)
“Man puts cotton swab too deep in his ear and accidentally returns himself to factory settings” might come across as a striking headline, but is only one of many examples of absurd headlines produced by the website De Speld. Since 2007, the website has provided Dutch citizens with their ‘Daily Standard for Reliable News’1. The website attracts over 1 million readers a month and is considered the largest satirical news outlet in the Netherlands (NRC, 2017). De Speld is comparable to other globally known news parody websites such as,
The Onion2, Der Postillon3, or Le Gorafi4.
News parody websites are classified as outlets with a clear intent to entertain and not to deceive (Verstraete, Bambauer & Bambauer, 2017). The editor-in-chief of The Onion notes that if someone is fooled by the content of the website, this should be considered “an
accident” (Verstraete et al., 2017, p.5). Jochem van den Berg, editor-in-chief of De Speld, acknowledges the potential ‘side-effects’ of (news) satire; however, he emphasises that De
Speld is simply ‘entertaining’ and diminishes the potential consequences (NRC, 2017).
Although entertainment is the primary driver behind news parody websites, the subjects – oftentimes social issues and political actors, are prone to criticism that implicitly influences the political attitudes of audiences. Therefore, scholars as well as journalists often refer to news parody as a serious matter that is changing public debate (Baym & Jones, 2012; Carlson & Peifer, 2013). Waisanen (2011, p.519) highlights the ‘ironic iconicity’ of news parody websites, which serves to “amplify moral judgments on politics and the media, playing an evaluative, satirical role in public discourse”.
Nevertheless, research to date has paid scarce attention to the processing and effects of news parody websites. The driving factor of conducting the present study is to help to address these research gaps. In addition, the results of the present study have clear societal
implications, since they offer insights into the ways in which a seemingly ‘harmless’ and comical website like De Speld can possibly affect citizens’ opinions and behaviours.
Generally, knowledge is an important factor when studying political satire, either as a prerequisite to process satirical content or as a result of exposure to satire. The current study focusses on the former: The effects of prior issue knowledge. Effect studies related to satire conclude that the attitudinal effects are not straightforward and largely dependent on the way satire is processed by the audience. Scholars conclude the interplay between prior knowledge, perceived humour, message resisting strategies, and the persuasiveness of satire for political attitudes is complex (e.g., Boukes, Boomgaarden, Moorman & de Vreese, 2015).
Next to prior topic knowledge, the processing and persuasiveness of political satire is affected by its tone. Satire is frequently classified within the dichotomy of horatian and juvenalian satire (Hill, 2013; Holbert, Hmielowski, Jain, Lather, & Morey, 2011). This experimental investigation includes these distinct types of humour. The interaction between one’s ability to process and the tone in a satirical piece, is an important factor when
determining the processing, and ultimately the effects of satirical content.
Besides the possible attitudinal effects of De Speld, the content of De Speld might affect political knowledge indirectly by stimulating behavioural intentions. Various scholars have concluded that exposure to TV-shows such as The Daily Show can be beneficial for citizens’ political knowledge (Young & Hoffman, 2012). The effects of written news parody websites are less likely to directly increase knowledge levels among their audiences.
However, written news parody websites might have an indirect effect on knowledge level by triggering further information seeking intentions (Feldman & Young, 2008). The following research question is central:
RQ: What are the effects of exposure to different forms of written news parody on
citizens’ (a) political attitude towards the satirized subject and (b) further information seeking intentions, and what role does prior knowledge play in this?
Theoretical Background
A New Form of Journalism: News Parody
News parody combines parody with elements of satire (Waisanen, 2011). Political satire is loosely described as any form of critique on society, policy, traditional news or journalism, often by making fun of a situation or a person (Buijzen & Valkenburg, 2004; Fife, 2016; Holbert, 2013). The methods used by news parody websites to criticize traditional news are twofold. On the one hand, articles of news parody websites are strongly related to written satire. For example, outlets such as De Speld or The Onion often criticise important current affairs, societal issues, and ridicule individual politicians. On the other hand, news parody indirectly criticises traditional journalism through mimicry. News parody evokes the impression of ‘real’ news by using similar formatting and website layouts as standard traditional news outlets (Tandoc, Lim & Ling, 2018). Both the Dutch website De Speld as well as the globally recognized website The Onion are ‘prima facie’ inseparable from traditional news websites in terms of style and presentation. In addition to similar
presentations, news parody thematically relates to ‘traditional’ news, given that both media types touch upon general societal themes such as politics or the economy (Berkowitz & Schwartz, 2016).
Scholars have emphasized that news parody is able to shed new light on societal issues and stimulate public dialogue uniquely with entertaining and humorous elements (Berkowitz & Schwartz, 2016; Fife, 2016). At times, traditional journalism fails to serve the public in this manner, because various ethical boundaries are imposed (Carlson & Peifer, 2013). The
implicit rules of ‘truth’ and ‘objectivity’ result in fact-based journalism that lacks interpretation (Meijer, 2001). Authors of news parody dissociate themselves from these ethical guidelines, which results in ‘news’ with critique (Waisanen, 2011). Therefore, we can determine that news parody is able to influence citizens’ attitudes with consideration a
number conditions number such as the type of humour and the ability of citizens to process the satire (Fife, 2016).
The Processing of Satirical Messages: Perceiving Satire as Funny
The current study uses several findings from pioneering studies that have examined the processing of satirical messages in order to explore the effects of news parody on citizens’ attitudes and behavioural intentions (Boukes et al., 2015; Holbert et al., 2011; LaMarre, Landreville, Young & Gilkerson, 2014; Nabi, Moyer-Gusé & Byrne, 2007).
Generally, studies conclude that satirical message processing is dependent on readers’ ability, and these findings are based on classic models of dual processing, such as the
Elaboration Likelihood Model of Persuasion from Petty and Cacioppo (1986) or the Heuristic Systematic Model from Chaiken & Trope (1999). The central route of processing is activated when the audience has the ability and motivation to process a message. Determining this ability and motivation can be accomplished by assessing citizens’ (political) knowledge. More specifically, ability relates to prior knowledge on a satirized subject, often retrieved through other outlets such as traditional news media (Boukes et al., 2015). Simply put, one asks if the audience is capable of discerning the satirical message in such a way that the joke is
understood. The motivation to process a satirical message is related to political interest (Holbert et al., 2011; Petty & Cacioppo, 1986). Since participants in the current study are deliberately exposed to a satirical piece, the motivational factor is of less importance here.
This research will follow the dichotomy of horatian and juvenalian satire. A
pioneering study on these specific forms of satire by Holbert and colleagues (2011) provides a better understanding of these distinct forms of humour. The study classifies horatian humour as “the truth with a smile”, whereas juvenalian satire is described as more “savage and merciless” (Holbert et al., 2011, p. 192). The tone of horatian satire is light-hearted and gentle, while the tone of juvenalian satire is ‘bitter’ and ‘acid’ (Hill, 2013). Horatian satire is
known as satire to ‘heal’, whereas juvenalian satire is referred to as satire to ‘wound’ (Holbert et al., 2011, p. 192). In summary, horatian satire relates closely to comedy, whereas
juvenalian satire has a stronger link with critique (Hill, 2013).
Previous studies show that juvenalian and horatian satire are processed differently based on the audience’s message processing ability (Holbert et al., 2011; LaMarre, 2014). The current study manipulates respondents’ ability to process a satirical message related to global warmth controversy: Participants are randomly assigned to either receive no information related to the satirized subject (low-ability) or are provided with a short piece of information related to the satirized subject (high-ability). The interaction between ability and humour is expected to affect perceived funniness – an influential factor of satire processing (Boukes et al., 2015; Holbert et al., 2011; Nabi et al., 2007).
The Political Satire Triad (PST) from Holbert (2016) helps us understand why the interaction between ability and humour type plays such a crucial part in satire processing. For satire to ‘work’ and to be perceived as funny, a level of agreement must exist between the satirist and the audience (Holbert, 2016). This mutual agreement consists of two parts: The worthiness of a subject to be satirized, and whether the nature of attack is legitimate (Holbert et al., 2011; Holbert, 2016). For the audience to determine this, some prior knowledge on the subject is required – especially when the audience is exposed to juvenalian satire. Simply put, while horatian satire is funny by its very nature, juvenalian satire can be considered as funny when audiences use existing knowledge structures. Therefore, it is expected that juvenalian satire will only be perceived as funny when the ability to process the satire is high: The notion that high-ability citizens perceive juvenalian satire as funnier, whereas low-ability citizens seem to appreciate horatian satire to a greater extent has been confirmed by previous studies (Holbert et al., 2011). Moreover, previous studies indicate that high-ability citizens are “turned-off” by the tone of horatian satire (Holbert et al., 2011, p. 202). Ability thus seems to
be the primary driver of determining the audience’s perception of humour in satire. Based on the findings discussed above, the following hypothesis is formulated:
H1: Participants in the high-ability condition who are exposed to an article with
juvenalian humour will perceive the article as funnier than participants in the low-ability condition (H1a), whereas participants in the low-ability conditions will perceive the article with horatian satire as funnier than participants in the high-ability condition (H1b).
The Persuasiveness of News Parody: Message-resisting Strategies
Political satire makes use of criticism, while simultaneously employing elements of persuasion (Holbert, 2013). In political satire, the satirist frames a politician, political issue, or societal issue in a certain (often negative) way. Therefore, various scholars explore the
persuasiveness of satirical messages in an attempt to measure satire’s influence on public opinion (e.g., Gruner, 1965). Persuasive attempts can lead to attitude change, but similar to other persuasive attempts, the audience often tends to resist messages that imply a certain persuasion (Fransen, Verlegh, Kiramani & Smit, 2015b; Knowles & Linn, 2004; Zuwerink Jacks & Cameron, 2003).
Several studies on the persuasiveness of satire indicate that when satire is perceived as funny, message resistance strategies are activated (Boukes et al., 2015; Holbert et al., 2011), specifically message discounting (LaMarre et al., 2014; Nabi et al., 2007). Message
discounting is defined as dismissing a message for a lack of valid information to base judgements upon (Nabi et al., 2007). It is a contesting strategy and can be interpreted as a distinct form of counterarguing (Wright, 1980). In the context of political satire, message discounting enables audiences to judge the content they are exposed to as a joke, rather than valuable information. In addition, humour type seems to directly activate message discounting strategies: Horatian satire is more actively discounted than juvenalian satire (LaMarre et al., 2014). This leads to the following hypothesis:
H2: Participants who perceive the satirical piece as funny will engage in higher levels
of message discounting than participants who perceive the satirical piece as less funny (H2a), and participants exposed to horatian satire will engage in higher levels of message
discounting than participants exposed to juvenalian satire (H2b).
As previously mentioned, De Speld has over one million readers a month, and is a well-known satirical news outlet in the Netherlands. Therefore, the current study should take into consideration that a large group of Dutch citizens are aware of the fact that De Speld is fictitious and intends to evoke humour from the audience. Questioning the credibility of a source is another message resistance strategy readers can engage in, and it is referred to as source derogation (Fransen et al., 2015b). Source derogation is dismissing a source as
untrustworthy and rejecting its validity (Wright, 1973; Zuwerink Jacks & Cameron, 2003). In the context of news parody, source derogation is closely related to media credibility and media literacy. Source derogation can be activated by simply recognising De Speld as a news parody website. However, to derogate the source, the audience requires some information on the source’s credibility and motivation. Scrutinizing and determining the purpose of certain media messages can be explained in terms of media literacy (Livingstone, 2004). In the context of the current study, it would mean that audiences are able to evaluate De Speld as a website whose purpose is entertainment. It is therefore expected that source derogation will only be prevalent among individuals already familiar with De Speld and understand how the website differs from traditional news sites. I therefore hypothesize:
H3: Participants who are familiar with De Speld will engage in higher levels of source
derogation than participants who are less familiar with De Speld.
The message-resisting strategy ‘source derogation’ is closely related ‘message discounting’, as both forms of resistance are contesting strategies, (Fransen et al., 2015a). A contesting resistance strategy is defined as a strategy where the source, content or persuasive
tactic is being scrutinized (Fransen, Smit & Verlegh, 2015a). Studies in the field of persuasive communication indicate that message content is challenged when the source of a message is derogated (Fransen et al., 2015a; Jenkins & Dragojevic, 2013): A source’s message is often rejected when the validity and credibility of a source are questioned. Therefore, the following hypothesis is formulated:
H4: Participants who engage in higher levels of source derogation will experience
higher levels of message discounting than participants who engage in lower levels of source derogation.
When audiences engage in message resisting strategies this will ultimately diminish the persuasiveness of messages (Fransen et al., 2015b; Knowles & Linn, 2004), which will be further examined in this study. As previously mentioned, satirists deliberately frame issues in a certain way, which may affect political beliefs. However, attitudinal effects are expected to reduce if audiences discount the satirical message as a joke intended for entertainment, not persuasion. In short, message discounting negatively affects satirist agreement, which in turn limits attitudinal effects. Therefore, the final hypothesis is proposed:
H5: Participants who engage in higher levels of message discounting will hold a less
negative attitudes towards the satirized subject than participants with lower levels of message discounting.
Seeking Further: The Gateway Mechanism
Previous studies indicate that the relationship between political satire exposure and information-seeking intention is positive (e.g. Boukes, 2018; Feldman & Young, 2008). For instance, recent work from Boukes (2018) shows that Google searches of the TTIP agreement dramatically increased following its coverage on a Dutch satirical TV-show. Another study shows that satirical TV-shows are often a gateway to traditional news outlets; commonly referred to as the gateway mechanism (Baum, 2003; Feldman & Young, 2008). Exposure to
political commentary in late-night talk shows “sparks the interest” to seek out information in traditional news outlets (Feldman & Young, 2008, p. 402). However, it is yet to be
determined if news parody websites such as De Speld provoke similar reactions as satirical TV shows regarding information-seeking intentions. It is expected that source derogation plays a role in this process as well. When the audience considers De Speld as less serious and just a humorous news outlet, less information seeking intention is expected. Hence, RQ1 is formulated.
RQ1: What is the relationship between exposure to different forms of news parody
and further information-seeking intentions and what role does source derogation play in this? Figure 1 gives a schematic overview of the formulated hypotheses and research question.
Method
Pilot Study
To make a justified decision on the satirized subject and to decide whether the manipulation of humour was executed correctly, an online pilot study was conducted (N = 20). Participants were recruited using a convenience sample; members of the
researcher’s own personal network were asked to participate. The sample of the pilot study consists of 9 men and 11 women, aged between 22 and 62 years (M = 30.15, SD = 13.30). First, participants were asked to indicate the importance of the following issues for them personally: 1) online privacy, 2) climate discussion and 3) vaccination of children. Afterwards, participants answered multiple statements related to these topics in order to assess issue polarization. When a political or social issue is polarized, citizens hold extreme opinions towards a specific topic (Fiorina & Abrams, 2008). It is unlikely that any new or contrasting information will affect their attitude. Therefore, topic selection is of great essence for the current study. The pilot study indicates that the articles related to online privacy and global warmth controversy are both suitable issues.
After topic selection, humour manipulation was assessed. A total of six articles were created: Using both a horatian and juvenalian article for each of the three aforementioned topics. Participants were exposed to either three horatian or three juvenalian articles. The results of the pilot study indicate that the articles related to global warmth controversy differ most from each other on harshness and critical tone. In combination with the relatively low levels of polarization of this issue, the articles from De Speld related to global warmth controversy were used in the experimental study. A full overview of the results is provided (see appendix II, tables 2 – 5).
Experimental Design
The proposed hypotheses were tested with an experimental 3 (horatian humour, juvenalian humour, no satire) x 2 (ability: high vs. low) between-subjects online design (see Table 1, Appendix I). Each participant was randomly assigned to one of the six conditions.
The data were collected between May 9 and 16, 2019. After participants were
informed about the goal of the study and gave their informed consent, three pre-test questions followed. Participants in the satire conditions filled out questions related to perceived humour, source derogation and message discounting. In each of the six conditions partipants filled out questions measuring their attitude and information seeking intentions. Next, participants were asked to provide socio-demographics. Finally, participants were exposed to a debriefing, explaining they were part of an experiment.
Participants
Participants were recruited with the help of ‘Respondenten.nl’. This ISO certified Market Research Company is specialized in recruiting respondents for quantitative and qualitative research. Participants received an incentive through the scoring system of Respondenten.nl. In total, 300 participants were recruited.
On average, participants needed 7.90 minutes to finish the experiment5. Some participants took over 30 minutes to finish the study. To ensure internal validity, these
participants were filtered out of the dataset6. Means on interview time per condition were used to filter out participants who finished the experiment in an unrealistic short time frame.
Participants who filled out the questionnaire in less than half of the average time were filtered out of the data set. The final sample consists of 276 participants.
Participants’ age ranged between 18 and 65 years (M = 45.43, SD = 12.95), and 52.5% were female. The sample consists of 19.1% lower educated, 38% middle educated and 42%
higher educated participants7. 32.5 % of participants identify as left-winged, 31.7% as middle-oriented, and 35.7% as right-winged8. Mean scores indicate that participants are slightly interested in politics (M = 3.53, SD = 1.10)9.
Stimulus Materials
Ability Manipulation
The current study manipulated respondents’ ability to process a satirical message. Prior to stimuli exposure, participants in the high-ability conditions (n = 132) were exposed to a 173-word non-opinionated piece of text related to global warmth controversy, which served to manipulate the participants’ ability. The information was retrieved through the website ‘Science Daily’ and was translated into Dutch10. In the low-ability conditions (n = 144), participants were not exposed to this information.
Humour Manipulation
Two versions of an article from De Speld related to global warmth controversy were used. The articles have a similar tenor, and both connect the ‘climate sceptic’ to a dying animal species. The Dutch party ‘Forum voor Democratie’ can be considered issue owner, in regard to global warmth controversy (Walgrave, Tresch & Lefevere, 2015). The articles directly refer to Thierry Baudet and Theo Hiddema, the party’s most prominent faces, and they consist of three short paragraphs, have the same layout and cover photo of Baudet in parliament. The author’s names are fictive. The manipulated articles are based on two published articles from De Speld11 but were adapted for the manipulated type of humour.
Horatian conditions. The horatian article is titled ‘Climate sceptic is dying out’. The
level of absurdity of this article is high. The article refers to ‘Thierry’ and ‘Theo’, the only two remaining climate sceptics in parliament, and claims that "the two are often sad on their own island", with a “great will to survive”.
Juvenalian conditions. The juvenalian article directly attacks the statements the two
politicians (particularly Baudet) make related to climate and global warming. The article criticises the ‘climate sceptic’ for not giving any substantive arguments and makes the cynical comparison that it is indeed hard to decide who to believe when it comes to global warming: “One politician with one opinion” or “more than 100 scientific studies that indicate the significant role of humankind in global warming”. In general, the article is more closely linked to critique, is more cynical and approaches Baudet’s person and ideas in a rough manner. Stimulus materials are provided in appendix III.
Measures
Pretest variables
Prior issue knowledge. In the pretest, participants were asked to indicate on a 11-point
scale how much they knew about the ‘climate discussion’. The answer scale ranged from -5 (very little) to +5 (very much), while scale values ranged from 0 to 11. While ability to process a satirical piece was manipulated with a small piece of information related to global warmth controversy, participants might already have a great level of knowledge on this topic. This question thus helps to categorize participants’ prior issue knowledge and was later used as a control variable. Results indicate that participants are well-informed regarding the climate discussion (M = 7.49, SD = 1.83).
Prior issue importance. Participants were asked: ‘How important do you think the climate discussion is?’, measured on a 11-point answer scale running from -5 (not at all
important) to +5 (very important). Again, scale values ranged from 0 to 11. Similar to prior
issue knowledge, issue importance is later used as a control variable. Results show that
Familiarity with De Speld. Participants were asked to indicate their level of familiarity
with the Dutch website De Speld. Four answer options were available: a) ‘I don’t know De
Speld’, b) ‘I have heard about De Speld but I don’t know what it is exactly’, c) ‘I know De Speld but I don’t know how the website differs from other news websites’, d) ‘I know De Speld and read their articles regularly, and I know how De Speld differentiates from other
news websites’. The statements were recoded using the following classification scheme: Answer a is classified as unfamiliar, b and c as a little familiar, and d as familiar. 38.4% of the sample were unfamiliar with De Speld, 31.9% knew it a little, and 29.7% were familiar with it.
Posttest variables
Perceived Funniness. How funny participants thought the article was to which they
were exposed was measured on a 11-point scale with one single statement (see Boukes et al., 2015). The scale ran from 0 (not funny at all) to 10 (very funny). Overall, participants did not perceive the articles as very funny (M = 4.58, SD = 2.75).
Source derogation. The scale to measure source derogation was inspired by the
trustworthiness dimension from the source credibility scale from Ohanian (1990; see also Nabi et al., 2007). The current study uses a shortened version adapted to De Speld.
Participants were asked to assess whether they thought the words ‘trustworthy’, ‘credible’ and ‘reliable’ were applicable to the Dutch website De Speld (5-point scale, ‘not at all applicable’ to ‘very applicable’). Items were reversed; higher scores on the scale mean higher levels of source derogation. Factor analysis with Varimax rotation indicates that the three items form a reliable scale (EV = 2.66, α = .934). The scale measuring source derogation runs from 1 to 5 (M = 3.61, SD = 1.04)
Message discounting. The level to which participants disregarded the message as not
were asked about their agreement with four 7-point Likert items: ‘The author of the message was just joking’, ‘The message was intended more to entertain than to persuade’, ‘The author was serious about advancing his views in the message (reversed)’, and ‘It would be easy to dismiss this message as simply a joke.’ Factor analysis with Varimax rotation indicates that the four statements can be combined into a reliable one-dimensional scale (EV = 2.60, α =.817). The computed scale measuring message discounting runs from 1 to 5 (M = 3.04, SD = 0.96).
Dependent variables
Attitude towards the satirized subject. In the current study attitude is related to satirist
agreement, the satirical piece of De Speld frames global warmth controversy in a negative way. Therefore, satirist agreement is high when participants hold negative attitude towards the ideas of climate sceptics. Two items were reversed so that a higher score on the scale
indicates a more negative attitude towards global warmth controversy. Attitude was measured using 4 statements that could each be answered on a 7-point scale (1 totally disagree to 7 totally agree). The statements were, ‘I think the role of humankind is negligibly small when it comes to global warmth’ (reversed), ‘Global warmth is a natural process’ (reversed), ‘I am very worried about global warmth’ and ‘I think reducing greenhouse gas emissions from consumers and businesses can help to slow down global warmth’. Results of a factor analysis indicate that the four statements can be combined into a one-dimensional scale (EV = 2.80, α = .854). The scale runs from 1 to 7 (M = 4.55, SD = 1.43).
Respondents in the experimental conditions were exposed to similar statements on a 5-point scale12. Results of a factor analysis with Varimax rotation indicate that the attitudinal statements related to global warmth controversy be combined into a one-dimensional scale as well (EV = 2.81, α = .850). The new scale measuring attitude runs from 1 to 5 (M = 3.44, SD = 0.99). To compare attitude towards global warmth controversy in the experimental
conditions with control conditions, both scales were standardized. The standardized scores are combined into one scale measuring attitude towards global warmth controversy for both experimental and control conditions. Standardized attitude scores range between -2.49 to 1.71 (M = 0.00, SD = 0.98)
Further Information Seeking. The second dependent variable asked for citizens’
intention to seek out further information on the subject global warmth controversy. The statements were inspired by Kaid (2002) and adjusted to the issue. The statements were, ‘Watch the news related to the issue for more information’, ‘Talk with friends about the issue’, ‘Read news articles related to the issue’, ‘Use the internet to find more information about the issue’. Statements could be answered on a 5-point scale (1 not likely, to 5 very likely). Factor analysis shows that the four statements participants in the experimental conditions were exposed to be combined into one scale, measuring intention to seek out further information in general (EV = 2.99, α =.887). The scale runs from 1 to 5 (M = 2.99, SD = 0.97). Participants in the control conditions were exposed to similar statements on the same 5-point scale. Factor analysis indicates that the four statements can be computed into a one-dimensional scale (EV = 2.55 α = .805). The scale runs from 1 to 5 (M = 3.26, SD = 0.91). A news scale measuring further information seeking intentions is created. The scale runs from 1 to 5 (M = 3.10, SD = 0.98).
Analyses
Randomization Check
Randomization was successful, and participants in the six conditions did not
significantly differ from each other in terms of age (F (5, 270) = 0.89, p = .488), gender (χ² (5) = 9.66, p = .086), educational level (F (5, 270) = 0.96, p = .444), political orientation (χ² (10) = 13.10, p = .218), political interest (F (2, 270) = 0.75, p = .588), familiarity with De
Speld (χ² (10) = 6.23, p = .769), prior issue knowledge (F(2, 270) = 1.04, p = .392), and prior
issue importance (F(5, 270)= 0.71, p = .616).
Manipulation Check
Humour manipulation was assessed by asking participants ‘What did you think of the tone of
the article?’. The question could be answered on a 7-point scale (1 ‘harsh’, 7 ‘light hearted’).
Participants exposed to a horatian article thought the article was more light hearted (M = 4.51,
SD = 1.66) than participants exposed to the juvenalian version (M = 3.75, SD = 1.39, t(177.1)
= 3.32, p <.001, 95% CI [0.31, 1.20], d = 0.50).
Hypotheses Testing
There were no significant main effects of ability (F (1,180) = 0.80, p = .371) or humour type (F (1,180) = 3.08, p = .081) on perceived funniness (see Table 6, appendix IV). High ability participants (M = 4.46, SD = 2.48) exposed to a juvenalian article did not
perceive the article as significantly funnier than low ability participants (M = 4.00, SD = 2.79), t (86) = 0.82, p = .414). Contrary to H1b, low ability participants (M = 4.31, SD = 2.85) exposed to a horatian article from De Speld perceive the article as less funny than high ability participants (M = 5.50, SD = 2.68), and this difference is significant (t(91) = -2.07, p = .041, 95% CI [-2.33, -0.05], d = 0.43). Results of a two-way ANOVA indicate a significant interaction-effect between the experimental factors (F(1, 178) = 4.21, p =.042, η² =.02; see Figure 2). Hypotheses 1a and 1b are rejected.
Results of an OLS regression indicate that humour type, ability, and perceived
funniness explained 13.2% of the variance in message discounting (R² = .13, F(3,178) = 9.02,
p <.001), and that perceived funniness is a significant predictor for message discounting: The
funnier participants found the De Speld article, the more they discounted the message as not containing any valid information (b* = 0.32, p <.001, 95% CI [0.06 0.16]). Hypothesis 2a is confirmed. An independent t-test indicates that the horatian article (M = 3.20, SD = 1.10) is more actively discounted by participants than the juvenalian article (M = 2.88, SD = 0.86),
t(180) = 2.22, p = .028, 95% CI [0.03, 0.58] , d = 0.33). Hypothesis 2b is confirmed.
A one-way ANOVA indicates that familiarity affects source derogation (F(2,179) = 17.44, p <.001, η² = .16). A post-hoc test with Bonferroni shows that participants familiar with De Speld engage in higher levels of source derogation than participants who are a little (Mdifference = 0.84, p <.001) or not at all familiar with De Speld (Mdifference = 0.98, p <.001). Participants who are a little familiar or not at all familiar with De Speld do not
Figure 2: Interaction-effect ability and humour
4 4,3 4,6 4,9 5,2 5,5 5,8 Horatian Juvenalian P er ce iv ed f u n n in ess
significantly differ from each other on levels of source derogation. Hypothesis 3 is confirmed. Figure 3 visualizes these findings.
Hypothesis 4 is tested with an OLS regression. Results indicate that source derogation explained 9.4% of the variance in message discounting (R² = .09, F(1,180) = 18.77, p <.001). The more participants derogate De Speld as a non-credible source, the more they discount the message as not containing any valid information (b* = 0.31, p <.001 95% CI [0.15, 0.41]). Hypotheses 4 is confirmed. Descriptive statistics for the experimental conditions are provided in Table 6 in appendix IV.
Hypothesis 5 expected participants who engage in higher levels of message
discounting to hold less negative attitudes towards the satirized subject. Results of an OLS regression indicate that hypothesis 5 is rejected: Message discounting is a weak and non-significant predictor for attitude towards the satirized subject, (R2 = .002, F(1,180) = 0.44, p = .51). An additional two-way ANOVA indicates that there are no significant differences in
2 2,4 2,8 3,2 3,6 4 4,4 4,8
None Little High
S o u rc e d er o g at io n
None Little High
between the six conditions regarding attitude towards global warmth controversy, F(2,270) = 0.45, p = .634. Descriptive statistics are provided in appendix IV in table 7.
The results of a two-way ANOVA indicate that there is no significant difference between the conditions and the intentions to seek out further information on the topic (F(2,270) = 0.46, p =.630). Descriptive statistics are provided in appendix IV in table 7. An additional OLS regression indicates that source derogation explained 5.7% of the variance for further information seeking intentions (R2 = .057, F(1,180) = 10.94, p <.001): The more participants derogate De Speld as an non-credible source, the less likely they are to seek out further information on global warmth controversy (b* = -0.24, p <.001, 95% CI [-0.36, -0.09]).
Exploratory Analysis
Next to hypotheses testing, the current study engaged in exploratory analyses. Results of a one-way ANOVA indicate that there is a significant effect of age on source derogation,
F(2,179) = 4.68, p = .010, η² = .05. Post-hoc testing with Bonferroni indicates that
participants younger Dutch audiences (18 – 35) engage in higher levels of source derogation than middle aged individuals (36 – 50), Mdifference= 0.52, p = .036, and older participants (51 – 65), Mdifference = 0.51, p = .016. Middle aged participants do not significantly differ from older participants in terms of source derogation.
Results of a OLS regression indicate that age is inversely correlated with levels of message discounting, F(1, 180) = 4.04, p = .046. The older the participant, the lower the levels of message discounting, (b* = -0.15, p = .046, 95% CI [-0.22, 0.00]).
The variables perceived funniness, source derogation, message discounting are combined into a correlation matrix with the two main dependent variables (see table 8, appendix IV13). Results indicate that perceived funniness positively correlates with message discounting, attitude and further information seeking, but negatively correlates with source
derogation. Moreover, further information seeking intentions are affected by message discounting and the attitude towards the satirized subject.
Discussion
The current study aimed to generate insights into the processing and persuasiveness of written news parody. The experimental design attempted to discern audiences’ responses to different forms of written news parody and the role of prior issue knowledge, effects were determined by zooming in on the processing of news parody.
The conditions under which audiences perceive satirical messages as funny directly contradict the findings from Holbert and colleagues (2011): High ability participants considered the horatian piece to be funnier than low ability participants. These clashing results should be interpreted in light of the context of the current study since they indicate that the processing of written news parody differs from other forms of political satire. The satirical cues in written news parody are distinct compared to other forms of satire (Rubin, Conroy, Chen & Cornwell, 2016). For example, news parody on TV is supported by a presenter who builds up to a joke, and is often supported by a laughing audience (Baym, 2005). These external humour cues cannot be found in written news parody – the only cue is the source itself. Moreover, written news parody lacks joke explanation, compared to other forms of satire. The prerequisite of prior topic knowledge to perceive written news parody as funny is highlighted with this finding.
It was expected the persuasiveness of the articles from De Speld would be largely affected by message resistance. Two crucial message-resisting strategies were expected to play a key role in the processing of written news parody: source derogation and message discounting. With regard to the former, the results of the current study indicate that familiarity with De Speld predicts source derogation: De Speld is considered a non-credible source when participants are familiar with their comic intent. Moreover, perceived funniness affects source
derogation, but in the opposite direction: When participants perceive the article as funny, the extent to which the source is derogated is lower. Studies in persuasive communication come to similar conclusions; humour positively influences source credibility (e.g. Chung & Zhoa, 2003). Thus, when audiences are aware that De Speld’s purpose is to evoke humour from their audiences, the website loses credibility. However, when De Speld succeeds in evoking a humorous reaction from the audience, the website’s credibility increases.
Discerning the article of De Speld as ‘just a joke’ is related to several factors. First, perceived funniness predicts message discounting. The laughter caused by an article from De
Speld activates a certain ‘discounting mechanism’, this finding is in line with those of
previous studies (LaMarre et al., 2014; Boukes et al., 2013; Holbert et al., 2011; Nabi et al., 2007). As hypothesised, message discounting is significantly higher when participants are exposed to a juvenalian article than a horatian article of De Speld, directly confirming
previous findings (LaMarre et al., 2014). Results confirm the notion that horatian satire has a stronger link with humour than juvenalian satire, whereas juvenalian satire is more closely linked to critique: Participants recognise the comical intent of horatian satire to a greater extent than juvenalian satire, and find it harder to determine whether juvenalian satire is meant to be a joke. When Dutch audiences come across an article from De Speld with a similar attacking tone the discounting cue is less likely to be activated, which could increase the persuasiveness of the article. Next to perceived humour and the tone of the satire, source derogation predicts message discounting: when De Speld is considered non-credible by the audience, the article provided by the website is considered of lesser value in relation to attitude formation, directly confirming previous studies (Fransen et al., 2015b; Jenkins & Dragojevic, 2013).
In summary, audience’s credibility judgements play a role in the discounting of the message. However, when audiences lack the ability to evaluate the source, as is the case when
they are unfamiliar with the comical intent of the website, message discounting can still be activated when an article from De Speld evokes humour. When the article is not perceived as funny, then it is expected the horatian article will be discounted to a greater extent than an article with a more attacking tone. Message discounting will be least prevalent among audiences who are exposed to a juvenalian article from a source unknown to them, that they will not perceive as funny.
Additional analysis indicates that age affects the degree in which the source is
derogated: Participants under the age of 35 engage in higher levels of source derogation than participants over 35. Moreover, as the participants’ age increases, levels of message
discounting decrease. These findings indicate that when older Dutch citizens come across an article from De Speld, it is assumed they will engage in message resisting strategies to a lesser extent. It is possible they consider the satirical news from De Speld as valuable information from a credible source, which in turn might affect their opinions, behaviour or attitudes. This seems to be a result of the fact that news satire is usually shared on online platforms (Rubin et al., 2016; Stevens & McIntyre, 2019). Older audiences might be less active on these platforms than younger individuals, which in turn leads to a difference in terms of media literacy; in essence, younger audiences are more skilled in recognising news parody compared to older audiences.
The stimuli ridicule the idea of global warmth controversy. It was expected that higher levels of message discounting would ultimately lead to a less negative attitude towards global warmth controversy, because persuasiveness of the satirical article would be lowered:
Audiences resist the persuasive attempt of the satirist. However, the results indicate that there are no differences between attitudes towards global warmth controversy regardless of the experimental conditions, which contradicts findings from persuasive communication, where message derogation predicts attitude (Jenkins & Dragojevic, 2011). When interpreting this
result, the topic of the articles should be taken into consideration: Participants think climate is an important topic, and feel they already know a lot about this issue. When topics are highly important to citizens, it is unlikely that exposure to one satirical piece would affect their attitude. This also explains the non-significant effects on further information seeking: If participants already know a great deal about the climate discussion, it is less likely that they will actively seek out more information related to this topic.
Limitations
The generalisability of these results is subject to certain limitations. The external validity of the experiment might be diminished because the stimuli materials were created by the researcher and not by professional satirists. The articles are distinct from articles De Speld usually publishes. The ‘juvenalian’ article, in particular, might be too harsh: Outside of this particular experiment, it is unlikely that audiences come across an article from De Speld with such an ‘attacking’ tone. Moreover, scholars have emphasised that news satire usually falls into the scope of horatian satire (Rubin et al., 2016). This assumption again highlights the notion that Dutch audiences will most likely never come across an article from De Speld with such explicit critique.
Second, a major limitation regarding internal validity should be noted. Initially, attitude towards the satirized subject was supposed to be measured with four statements on a 7-point scale. While programming the survey, an error occurred which exposed control conditions to a 5-point scale. Scores were standardized to compare the attitudinal differences. However, it did result in the fact that experimental conditions were able to answer the
statements on a wider scale than control conditions, which gave experimental conditions the opportunity to answer the statements in a more nuanced way than control conditions (Joshi, Kale, Chandel & Pal, 2015).
Research Implications
The current study set the stage for other scholars to engage in experimental
investigations related to news parody. The present study tried to establish whether political attitudes could be affected by news parody, but could not determine whether news parody is able to affect attitudes towards individual politicians. Therefore, I highly encourage other scholars to investigate the effects of news parody exposure on candidate evaluations. Previous study related to other forms of satire have established this effect (e.g. Baumgartner, 2008).
Next to engaging in experimental studies, I invite other scholars to follow the example by Boukes (2018) and use digital data to determine the relationship between news parody exposure and information seeking behaviour beyond self-reports and behavioural intentions.
Conclusion
To conclude, the findings suggest an interplay between variables that is complex and
reciprocal. Similar to other studies related to other forms of political satire, the effects of news parody are not straightforward. Ultimately, attitudinal and behavioural effects of written news parody are limited. However, this might be a result of topic choice as discussed earlier. The findings of the current study clearly suggest message resisting strategies are less prevalent under certain conditions and among individuals above a certain age. This highlights the notion that the possible persuasive effects of news parody cannot be minimized to only "an
accident". Findings hint in the direction persuasive effects of news parody websites should not be diminished, contrary to the assumption of news parody authors: The articles provided by
De Speld are not always considered a joke by an entertainment-based website. The current
study contributes to an extension of studies related to the effects of political entertainment, as it concludes how the processing of news parody differentiates from other forms of satire. The saying “laughter is the best medicine” is found to be true in the current study, as perceived funniness seems the best way to overcome the potential ‘side-effects’ of news parody.
References
Baum, M. A. (2003). Soft news goes to war: Public opinion and American foreign policy in
the new media age. NJ: Princeton University Press.
Baumgartner, J. C. (2008). Polls and elections: Editorial cartoons 2.0: The effects of digital political satire on presidential candidate evaluations. Presidential Studies
Quarterly, 38(4), 735-758. doi:10.1111/j.1741-5705.2008.02675
Baym, G. (2005). The Daily Show: Discursive integration and the reinvention of political journalism. Political Communication, 22(3), 259-276.
doi:10.1080/10584600591006492
Baym, G., & Jones, J. P. (2012). News parody in global perspective: Politics, power, and resistance. Popular Communication, 10(1-2), 2-13.
doi:10.1080/15405702.2012.638566
Berkowitz, D., & Schwartz, D. A. (2016). Miley, CNN and The Onion: When fake news becomes realer than real. Journalism Practice, 10(1), 1-17.
doi:10.1080/17512786.2015.1006933
Boukes, M. (2018). Agenda-setting with satire: How political satire increased TTIP’s saliency on the public, media, and political agenda. Political Communication. doi:10.1080/10584609.2018.1498816
Boukes, M., Boomgaarden, H. G., Moorman, M., & De Vreese, C. H. (2015). At odds: Laughing and thinking? The appreciation, processing, and persuasiveness of political satire. Journal of Communication, 65(5), 721-744. doi:10.1111/jcom.12173
Buijzen, M., & Valkenburg, P. M. (2004). Developing a typology of humor in audiovisual media. Media psychology, 6(2), 147-167. doi:10.1207/s1532785xmep0602_2 Carlson, M., & Peifer, J. T. (2013). The impudence of being earnest: Jon Stewart and the
doi:10.1111/jcom.12019
Chaiken, S., & Trope, Y. (Eds.). (1999). Dual-process theories in social psychology. New York: NY: Guilford Press.
Chung, H., & Zhao, X. (2003). Humour effect on memory and attitude: Moderating role of product involvement. International Journal of Advertising, 22(1), 117-144.
doi:10.1080/02650487.2003.11072842
Fiorina, M. P., & Abrams, S. J. (2008). Political polarization in the American public. Annual
Review Political Science, 11(1), 563-588.
doi:10.1146/annurev.polisci.11.053106.153836
Fransen, M. L., Smit, E. G., & Verlegh, P. W. (2015a). Strategies and motives for resistance to persuasion: An integrative framework. Frontiers in Psychology, 6, [1201].
doi:10.3389/fpsyg.2015.01201
Fransen, M. L., Verlegh, P. W., Kirmani, A., & Smit, E. G. (2015b). A typology of consumer strategies for resisting advertising, and a review of mechanisms for countering them.
International Journal of Advertising, 34(1), 6-16.
doi:10.1080/02650487.2014.995284
Feldman, L., & Young, D. G. (2008). Late-night comedy as a gateway to traditional news: An analysis of time trends in news attention among late-night comedy viewers during the 2004 presidential primaries. Political Communication, 25(4), 401-422.
doi:10.1080/10584600802427013
Fife, J. (2016). Peeling The Onion: Satire and the complexity of audience response. Rhetoric
Review, 35(4), 322-334. doi:10.1080/07350198.2016.1215000
Gruner, C. R. (1965). An experimental study of satire as persuasion. Communications
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. New York, NY: Guilford Publications.
Hill, M. R. (2013). Developing a normative approach to political satire: A critical
perspective. International Journal of Communication, 7(Breaking Boundaries), 324-337. Retrieved from http://ijoc.org/index.php/ijoc/article/ view/1934/856
Holbert, R. (2013). Developing a normative approach to political satire: An empirical perspective. International Journal Of Communication, 7(Breaking Boundaries), 305– 323. Retrieved from https://ijoc.org/index.php/ijoc/article/view/1933
Holbert, R. L. (2016). Entertainment television and political campaigns: The political satire appropriateness (PSA) model. Praeger Handbook of Political Campaigning in the
United States [2 volumes], 171.
Holbert, R. L., Hmielowski, J., Jain, P., Lather, J., & Morey, A. (2011). Adding nuance to the study of political humor effects: Experimental research on juvenalian satire versus horatian satire. American Behavioral Scientist, 55(3), 187-211.
doi:0.1177/0002764210392156
Jenkins, M., & Dragojevic, M. (2013). Explaining the process of resistance to persuasion: A politeness theory-based approach. Communication Research, 40(4), 559-590.
doi:10.1177/0093650211420136
Joshi, A., Kale, S., Chandel, S., & Pal, D. K. (2015). Likert scale: Explored and explained. British Journal of Applied Science & Technology, 7(4), 396. doi:10.9734/BJAST/2015/14975
Kaid, L. L. (2002). Political advertising and information seeking: Comparing exposure via traditional and Internet channels. Journal of Advertising, 31(1), 27-35.
Knowles, E. S., & Linn, J. A. (2004). The importance of resistance to persuasion. In E. S. Knowles & J. A. Linn (Eds.), Resistance and Persuasion. Mahwah, New Jersey: Lawrence Erlbaum Associates.
LaMarre, H. L., Landreville, K. D., Young, D., & Gilkerson, N. (2014). Humor works in funny ways: Examining satirical tone as a key determinant in political humor message processing. Mass Communication and Society, 17(3), 400-423.
doi:10.1080/15205436.2014.891137
Livingstone, S. (2004). Media literacy and the challenge of new information and communication technologies. The communication review, 7(1), 3-14. doi:10.1080/10714420490280152
Meijer, I. C. (2001). The public quality of popular journalism: Developing a normative framework. Journalism Studies, 2(2), 189-205. doi: 10.1080/14616700120042079 Nabi, R. L., Moyer-Gusé, E., & Byrne, S. (2007). All joking aside: A serious investigation
into the persuasive effect of funny social issue messages. Communication
Monographs, 74(1), 29-54. doi:10.1080/03637750701196896
NRC. (2017, 20 december). ‘We nemen politici niet meer serieus en gaan vanwege komieken
naar de stembus’. Retrieved from:
https://www.nrc.nl/nieuws/2017/12/20/we- nemen-politici-niet-meer-serieus-en-gaan-vanwege-komieken-naar-de-stembus-a1585668
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers' perceived expertise, trustworthiness, and attractiveness. Journal of advertising, 19(3), 39-52. doi:10.1080/00913367.1990.10673191
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion:
Central and peripheral routes to attitude change. New York, NY: Springer-Verlag.
scholarly definitions. Digital Journalism, 6(2), 137-153. doi:10.1080/21670811.2017.1360143
Rubin, V. L., Conroy, N. J., Chen, Y., & Cornwell, S. (2016). Fake news or truth? Using satirical cues to detect potentially misleading news. In Proceedings of NAACL-HLT (pp. 7–17). doi:10.18653/v1/W16-0802
Stevens, E. M., & McIntyre, K. (2019). The layers of the onion: The impact of satirical news on affect and online sharing behaviors. Electronic News, 13(2), 78-92.
doi:10.1177/1931243119850264
Verstraete, M., Bambauer, D.E., & Bambauer, J. R. (2017). Identifying
and countering fake news. Arizona Legal Studies Discussion Paper, 17(15), 1–33. doi:10.2139/ssrn.3007971
Walgrave, S., Tresch, A., & Lefevere, J. (2015). The conceptualisation and measurement of issue ownership. West European Politics, 38(4), 778-796.
doi:10.1080/01402382.2015.1039381
Waisanen, D. J. (2011). Crafting hyperreal spaces for comic insights: The Onion news network's ironic iconicity. Communication Quarterly, 59(5), 508-528.
doi:10.1080/01463373.2011.615690
Wright, P. L. (1973). The cognitive processes mediating acceptance of advertising. Journal of
marketing research, 10(1), 53-62. doi:10.1177/002224377301000108
Wright, P. L. (1980). Message-evoked thoughts: Persuasion research using thought
verbalizations. Journal of Consumer Research, 7(2), 151-175. doi:10.1086/208804. Young, D. G., & Hoffman, L. (2012). Acquisition of current-events knowledge from political
satire programming: An experimental approach. Atlantic Journal of
Zuwerink Jacks, J., & Cameron, K. A. (2003). Strategies for resisting persuasion. Basic and
applied social psychology, 25(2), 145-161. doi:10.1207/S15324834BASP2502
Notes
1 The slogan of the website De Speld is ‘Uw Vaste Prik voor Betrouwbaar Nieuws’, which loosely
translates to ‘Your Daily Standard for Reliable News’.
2 The Onion is considered the largest satirical news website in the United States. Recently Stevens &
McIntyre (2019) published a study related to the behavioural effects of this satirical news website. The study
approaches the effects in a different manner, but is indicates how studies related to the effects of news parody
websites gain attention in the communication science literature.
3 Der Postilion is the largest satirical news website in Germany, and is accessible through the following
link: https://www.der-postillon.com/
4 Le Gorafi is a French news parody website, and is accessible through the following link:
http://www.legorafi.fr/
5 Participants were asked if they faced any problems during the experiment. None of the participants
indicated that they could not read the stimuli materials. Therefore, participants are only filtered out based on
interview time.
6. The experiment was expected to take 10 minutes maximum, depending on condition. Therefore, 30
minutes is an unrealistically long timeframe. It is very likely these participants engaged in other activities while
being part of the experiment, which lowers internal validity. Other factors than the stimuli materials may have
affected these participants.
7 Participants who finished elementary school, or finished high-school on the level ‘VMBO’ or
‘MAVO’ are classified as lower-educated. Participants with a degree in secondary vocational education (MBO) or finished high-school on the level ‘HAVO’ or ‘VWO’ are classified as middle-educated. Participants with a
degree in a university of applied sciences (HBO) or a degree in a university (WO) are classified as ‘higher-educated’.
8 Participants could indicate their political position on a 10-point scale. The following classification
scheme is used to categorize participants’ political preference: Left-winged (values 1 to 4), Middle-oriented (5), Right-Winged (values 6 to 10).
9 Political interest was measured on a 7-point scale running from (1) not interested to (7) very interested.
10 Ability manipulation is retrievable through the following link
https://www.sciencedaily.com/terms/global_warming_controversy.htm
11 Original articles from De Speld: https://speld.nl/2014/03/04/klimaatsceptici-sterven-uit/ and
https://speld.nl/2019/01/09/wie-te-geloven-deze-101-rapporten-over-klimaatverandering-of-drie-uit-de-context-getrokken-grafieken-van-thierry-baudet/
12 Initially, participants in the control conditions should have been able to answer the statements on a
similar scale as respondents in the experimental conditions. However, an error in the questionnaire programming
led to participants in the control conditions being presented with a 5-point scale in contrast to 7-point scale. For
this reason, scores are standardized and later computed into one scale.
13 Model testing with the PROCESS tool from Andrew Hayes (2013) showed no indirect relationships
between the independent variables, potential mediators, and dependent variables in the design. Therefore, the
Appendix I: Experimental Design
Table 1: Experimental conditions
Horatian Satire Juvenalian Satire No Satire
No prior information A (n = 46) B (n = 45) F* (n = 41)
Prior information provided
C (n = 48) D (n = 48) E* (n = 48) Note. N = 276
Appendix II: Results of the Pilot Study
Table 2: Results of pilot study: topic importance
Topic M SD
1. Global warmth controversy 3.10 2.05
2. Online Privacy 2.90 2.38
3. Vaccination of Children 3.10 2.47
Note. Participants were asked: ‘Can you please indicate how important [topic] is to you?’. Answer scale ran from -5 (not important at all) +5 (very important).
Table 3: Results of pilot study: articles related to global warmth controversy
Statement MD t df p SE
The article of De Speld
… is serious - 0.20 - 0.26 13.79 .799 0.77
… makes me angry 0.50 0.80 18 .435 0.63
… is funny - 0.80 - 1.04 18 .315 0.77
… is critical - 1.70 - 3.94 18 .003 0.49
… is light hearted 0.10 0.13 18 .896 0.75
… has a cynical tone - 0.90 - 0.87 18 .387 0.78
… is harsh - 1.50 - 2.16 18 .045 0.69
Note. Participants were asked to indicate how much they (dis)agreed with the statements on a 7-point scale.
Independent t-test was conducted with using the horatian article and the juvenalian article as factors
Table 4: Results of pilot study: articles related to online privacy
Statement MD t df p SE
The article of De Speld
… is serious - 0.70 - 0.80 13.65 .436 0.87
… makes me angry 0.30 0.58 18 .572 0.52
… is funny - 0.10 - 0.16 18 .878 0.64
… is critical 0.30 0.36 18 .723 0.83
… is light hearted 1.0 1.37 18 .187 0.73
… has a cynical tone - 0.50 - 0.61 18 .547 0.81
… is harsh 0.10 0.15 18 .881 0.66
Note. Participants were asked to indicate how much they (dis)agreed with the statements on a 7-point scale.
Independent t-test was conducted with using the horatian article and the juvenalian article as factors
Table 5: Results of pilot study: articles related to vaccination of children
Statement MD t df p SE
The article of De Speld
… is serious - 0.60 - 0.79 18 .438 0.76
… makes me angry 0.90 1.16 18 .263 0.78
… is funny 1.00 1.87 18 .078 0.54
… is critical - 0.10 - 0.14 18 .888 0.70
… is light hearted 0.50 0.72 14.21 .056 0.75
… has a cynical tone 0.40 0.54 18 .598 0.75
… is harsh 0.50 0.72 14.75 .482 0.70
Note. Participants were asked to indicate how much they (dis)agreed with the statements on a 7-point scale.
Independent t-test was conducted with using the horatian article and the juvenalian article as factors
Appendix III: Stimulus Materials
Appendix IV: Results of Experimental Study
Note. Perceived funniness was measured on a 11-point scale. Participants were asked “How funny was the article of De Speld you just read related to global warmth controversy?”. Source derogation was measured on a 5-point scale by asking participants do indicate if the words ‘credible’, ‘trustworthy’ and ‘reliable’ were applicable for De Speld. Message discounting was measured by four statements on a 7-point scale indicating the level in which participants were able to judge the content as a joke.
Table 6: Descriptive statistics experimental conditions: perceived funniness, source derogation, message discounting
Variable Perceived Funniness Source Derogation Message Discounting
Ability Humour M SD M SD M SD N Low Horatian 4.31 2.85 3.89 1.13 3.27 1.07 45 Juvenalian 4.46 2.48 3.60 1.07 2.91 0.86 41 Total 4.38 2.67 3.75 1.11 3.10 0.98 86 High Horatian 5.50 2.68 3.52 1.00 3.13 0.98 48 Juvenalian 4.00 2.79 3.44 0.96 2.86 0.87 48 Total 4.75 2.83 3.48 0.97 2.99 0.93 96 Total Horatian 4.92 2.81 3.70 1.08 3.20 1.02 93 Juvenalian 4.21 2.65 3.51 1.00 2.88 0.86 89 Total 4.58 2.75 3.60 1.04 3.04 0.96 182
Table 7: Descriptive statistics all conditions: attitude and further information seeking intentions
Variable Attitude Further information Seeking
Ability Humour M SD M SD N Low Horatian -0.01 0.95 3.23 0.92 46 Juvenalian 0.01 0.88 2.94 1.12 45 No Satire -0.05 1.05 2.91 1.02 41 Total -0.01 0.96 3.03 1.03 132 High Horatian 0.01 1.04 3.23 0.91 48 Juvenalian 0.05 1.03 3.21 0.93 48 No Satire -0.10 1.05 3.00 0.90 48 Total 0.01 1.04 2.99 0.91 144 Total Horatian 0.00 1.00 3.23 0.91 94 Juvenalian 0.08 0.96 3.08 1.03 93 No Satire -0.08 1.04 2.96 0.95 89 Total 0.00 1.00 3.09 0.97 276
Note. Attitude scores are standardized.
Table 8: Means, standard deviations, Pearson correlation for continuous variables (n = 182) Mean SD 1 2 3 4 5 1. Perceived funniness 4.58 2.75 1 2. Source derogation 3.60 1.04 -.13* 1 - 3. Message discounting 3.04 0.96 .33* .31** 1 4. Attitude 4.55 1.42 .14* -.05 .05 1 5. Further information seeking 3.02 0.99 .14* -.24* -.10 -.31** 1
All values in bold are significant
*Significant one-tailed correlations
