The re-emergence of party press in Denmark:
As credible as traditional media or a wolf in sheep’s clothing?
Written by: Sanni Jensen Student ID: 12367788
Master’s Thesis
Graduate School of Communication Master’s programme Communication Science
Supervised by Sabine Geers Handed in on June 3rd 2019
Abstract
In Denmark, several political parties have re-introduced an old form of media: The party press. Using tools formerly reserved for journalists, they now create news content on own, self-funded media platforms that resemble the old party press system. Using both right-wing
and left-wing party press, a survey experiment is carried out to test whether citizens recognize the new party press as such and to investigate whether they think news content
from the new party press is less credible than content from traditional media. The results show that citizens most often do not recognize party press, and they do not rate party press
content differently than traditional media on credibility. However, if citizens recognize the source of the content as party press, they rate the credibility of the content significantly lower
than content from traditional media. The implications of these findings are discussed, and possible future research options are proposed.
Table of Contents
INTRODUCTION ... 1
THEORY ... 2
THE SCANDINAVIAN PARTY PRESS ... 2
MEASURING CREDIBILITY IN NEWS ... 4
METHOD ... 7 STUDY DESIGN ... 7 SAMPLE ... 7 PROCEDURE ... 8 STIMULUS MATERIAL ... 8 MANIPULATION CHECKS ... 9 MEASURES ... 9 ANALYSIS ... 10 RESULTS ... 11 DISCUSSION ... 14 REFERENCES ... 18
APPENDIX 1: FULL SURVEY EXPERIMENT ... 24
APPENDIX 2: LEFT-WING PARTY PRESS STIMULI ... 29
APPENDIX 3: RIGHT-WING PARTY PRESS STIMULI ... 31
Introduction
Within the last few years, the majority of Danish political parties have created their own media platforms. Using the toolbox of journalists, they now produce articles for their own online news sites, weekly news podcasts and online-TV. At live press conferences a green microphone with the logo of the Conservative People’s Party now stands out between the traditional media’s, and well-known TV news hosts are paid by parties to do interviews with their leaders.
Some of the parties claim that their media outlets are completely independent from the parties and free to cover anything they want, just like traditional media (Lessel, 2018; Rohde & Jessen, 2018), while others are applying for government media subsidies that are normally reserved for independent media (Langberg, 2018; “Ministry of Culture”, 2017). Because these new outlets are borrowing the legitimacy of traditional media, and
communicate outside their own group of members, they are not only an extension of the parties’ regular communication. But because the owners of, and the leadership behind, these new platforms are obviously partisan, they do also not just appear to be a new type of
independent media. In fact, what it arguably resembles the most is a fusion of party
communication and traditional news media that has not been seen for many decades in the country: The party press.
The re-emergence of party press in Denmark is an unstudied new trend that adds to the growing possibilities of not noticing the difference between independency and
partisanship in media (Freeman & Spyridakis, 2004; Kiousis, 2006; McGrew, Breakstone, Ortega, Smith & Wineburg, 2018). Many party press outlets claim to be independent, but they endorse a certain party, both behind the scenes and explicitly in the content (Jensen, 2018). At the same time, research shows that a confirmation bias effects the way citizens evaluate information, meaning that information that is more aligned with our own beliefs is perceived as being more credible (Alsaad, Taamneh & Al-Jedaiah, 2018; Giner-Sorolla & Chaiken, 1994; Kim, 2019). Because of this, it is interesting to study how the Danish citizens perceive the neutrality of party press and to what extent their bias towards parties and the media influence this. To study this, the following research question is asked:
“To what extent do Danish citizens perceive new forms of party press as credible compared to traditional media?”
Because party press in this new context is still an unstudied trend, another explorative sub-research question is asked:
“To what extent are Danish citizens able to recognise the new form of party press for what it is?”
First of all, the theoretical and historical backgrounds of party press are discussed and compared to the new, re-emerged party press. Then, a broader framework of theory on confirmation bias and credibility of news media is presented. The method is then discussed and described before the results are presented. Lastly, a discussion of the results in light of the theory is put forward and ideas for future studies are presented.
Theory
The Scandinavian party press
In Denmark, the party press originally emerged from the 1830’s to the 1850’s. It grew steadily for about a hundred years until the omnibus newspapers took over from the 1960’s (Hjarvard, 1995; Schultz, 2007). Before this, each political party had a newspaper
representing and openly supporting them in each city (Gregersen, 2018). The topics that were covered and opinions expressed always fit the party’s views. In general, the literature talks about how party press did not separate “news from views”, meaning that there was no clear distinction between content expressing opinions and reports of real facts (Hjarvard, 2007).
After the Second World War other characteristics of newspapers were more valued, and soon the “omnibus” press dominated in Denmark. This was characterized by standards of objectivity and detachment from political affiliations (Hjarvard, 2007). Now it was more important to inform people than to spread opinions (Allern, 2007; Gregersen, 2018; Hjarvard, 2007).
In recent years, the idea of a segmented press has started to rise (Schultz, 2007). The idea is that news media, in order to keep being relevant in an increasingly difficult market, target their content to certain groups of people and not to everyone as with the omnibus press. At the same time, the internet and especially social media has made it
possible to create and spread content in a different, quicker way. Information, misinformation, fake news, biased news and opinions are all mixed together online, and citizens have to navigate through it to uncover what is credible and what is not (Fletcher & Park, 2017; McGrew, Breakstone, Ortega, Smith & Wineburg, 2018; McGrew, Ortega, Breakstone &
Wineburg, 2017). Blumler (2018) argues that what we now have is a ‘communication abundance’, and that it is getting increasingly difficult to win the audience’s attention.
As a result, political communicators have for a long time been using media logic to get the media’s and the public’s attention (Blumler & Kavanagh, 1999; Hepp, Hjarvard & Lundby, 2015; Strömbäck, 2008). This has caused an increased mediatization of the political sphere (Couldry & Hepp, 2013; Hjarvard, 2008). Arguably, this mediatization created an environment for the re-emergence of party press, because political parties learned so much about how the news media work that they might as well create their own (Jensen, 2018). As a result, several of these partisan media outlets have emerged (Alternativet, n.d.; Dansk Folkeparti, n.d.; Enhedslisten, n.d.; Netavisen Pio, n.d.). Just like the historic party press, the new party press outlets mix news with their own views, clearly advocating for a certain way to view the world (Borre & Rohde, 2018b; Jensen, 2018).
However, there are of course some differences between old and new party press. Whereas old party press was clearly linked to a certain party, some political parties today do not want to get associated with their party media. For instance, the editor of the party press outlet associated with The Social Democrats, Netavisen Pio, do not think the ties between the party and the media are too strong for it to be considered independent (Borre & Rohde, 2018a), even though independent news media have uncovered numerous close links between them (Andersen, 2018; Andersen & Geist, 2018; Borre & Rohde, 2018b). On the other end of the scale, The Danish People’s Party’s Dit Overblik is openly founded and funded by the party, and the editor-in-chief has said that the party will fire him if he does not follow the line the party has laid out (Nielsen, 2018). These differences are interesting because they make the new party press different from the old where all media openly stated where they belonged (Gregersen, 2018). Another difference between old and new party press is that you no longer need to actively buy the newspaper in order to get exposed to it. Today, party press exists online and are more easily accessed by everyone and shared with your network.
However, the similarities are just as striking. For instance, using several definitions, Allern and Blach-Ørsten (2011) argued that the “old” party press was defined by the political loyalty the newspapers showed their specific parties and that the links of ownership, content and readership between the newspapers and the parties mattered. In a definition of “old party press by Høyer (2005) this taken into account when he states that a “a ‘party paper’ is a newspaper owned, staffed and directed by a political party or by political party affiliations” (p. 76). This definition can almost be directly transferred to the political party news media
being launched today. Only the word “newspaper” needs to be changed into e.g. “news outlet” so that it also captures the online news landscape that is today.
Measuring credibility in news
Because so many outside the traditional news media industry are now using the toolboxes of journalists, and thereby borrowing legitimacy from news media that is striving to be impartial, the audiences are now often required to assess whether a news source is credible or not. When people are assessing the credibility of a source, multiple factors have shown to have an impact. In this section, credibility measures in general will be discussed first, then the effects on credibility of both the media and the content will be presented, and lastly the characteristics of the individual that have shown to have an impact (Dalton, Beck & Huckfeldt, 1998; Golan, 2010; Lee, 2010; Meyer, 1988; Westerwick, Kleinman & Knobloch-Westerwick, 2013).
In one of the first studies that investigated how to measure credibility, Gaziano and McGrath (1986) identified 12 credibility items. Whether a news piece was: Fair, unbiased, told the whole story, accurate, respected people’s privacy, watched out for people’s interests, was concerned about the community’s well-being, if it separated fact and opinion, could be trusted, was concerned about the public interest, was factual and was made by well-trained reporters (Gaziano & McGrath, 1986, p. 454). These items are still used today in more or less moderated versions (see e.g. Đorđević, 2019; Engelke, Hase & Wintterlin, 2019; Wang & Hickerson, 2016).
Using credibility measures, studies have not surprisingly found that credibility of content depends on the perceived credibility of the messenger – meaning both the sources in the content and the platform on which the content is published (Oydeji, 2010;
Kalogeropoulos, 2018; Roberts, 2010). In the latest annual Reuters Digital News Report, UK citizens for instance perceived BBC as a lot more trustworthy than Daily Mirror, and Danish citizens found their national broadcasting service, DR, more trustworthy than a big tabloid newspaper, Ekstra Bladet (Kalogeropoulos, 2018).
Looking at trust in the messenger versus the message, Cheng and Lo (2012) tested how different news formats were evaluated differently in terms of credibility across different news organizations. What they found was that the effect of the different news formats was overrun by the effect of the perceived credibility of the news organization, meaning that the effect of the formats did not mean as much as the organization itself. Interestingly, Roberts (2010) used the Meyer (1988) and Flanagin and Metzger (2000) credibility scales to try and measure both message and messenger credibility at the same
time. In his experimental survey study, participants evaluated both types of credibility of articles from well-known newspapers that were accompanied by a lot or very little information on how the story had been made. The results showed that the two credibility scales that measured message and messenger respectively to a great extent measured the same thing. However, according to Roberts (2010) there seemed to be enough distinction to take both into consideration. These findings are interesting in a party press context because the effect of the news media itself (i.e. the party press) might show to be more important than the actual news they are producing in terms of credibility.
Furthermore, studies have indicated that the more a person uses a media outlet, the more likely they are to recognize it and rate it as being credible (Johnson & Kaye, 2000; Rimmer & Weaver, 1987; Wayne & Hu, 1994). This means that a new media outlet – party press or not – might be expected to be evaluated as less credible simply because few people are familiar with it. Furthermore, identifying the source can be more difficult and blurred in regard to online media than traditional media (Cheng & Lo, 2012; Freeman & Spyridakis, 2004; Kiousis, 2006; Trench & Quinn, 2003). This means that people can be expected to rate party press content as less credible because the party press outlets are relatively new and therefore arguably more unknown that traditional media. However, if people do not recognize these online sources for what they are the effect might be smaller, and therefore the following hypotheses are put forward:
H1: People perceive party press content as less credible than content form traditional media.
H2: People find party press content less credible if they recognize the party press’ affiliation with a political party.
Furthermore, research has shown that people have a confirmation bias, which means that they interpret, search for and evaluate information in a way that supports what they already know or believe (Alsaad, Taamneh & Al-Jedaiah, 2018; Giner-Sorolla &
Chaiken, 1994; Kim, 2019; Klayman, 1995; Nickerson, 1997). Meyer, Marchionni & Thorson (2010) for instance found that agreeing with the author was a strong predictor of perceived expertise of journalists. In an experimental setting, having the same opinion as the one the story supported highly predicted the perceived expertise of the author. Studying this
confirmation bias in news content on same-sex marriage from mainstream media and blogs, Kim (2014) found that those who did not agree with the content’s stance on same-sex
marriage perceived both the mainstream media and the blog post as less credible than those who agreed with the content’s stance on the issue (Kim, 2014, p. 29). In a study of pro-Palestinian’s and pro-Israeli’s evaluation of the bias in news coverage of the conflict, both sides found the news coverage to be biased in favour of the opposite group, whereas a more neutral third group perceived the news coverage as less biased than the two partisan groups (Giner-Sorolla & Chaiken, 1994). This confirmation bias means that those who agree with the political party that produces the party press content might be expected to rate it as more credible than those who do not.
Furthermore, literature show that people with strong partisanship in any direction are more likely to perceive the media as being biased (Dalton, Beck & Huckfeldt, 1998; Giner-Sorolla & Chaiken, 1994). This means that it may be expected that there is a difference between a confirmation bias among those with moderate political affiliations and partisanship among those who are more extremely partisan in any direction. Lee (2005) saw this in his study of factors influencing media bias perception. Using data from two surveys with a total of more than 5,000 respondents, he found that the stronger general partisanships respondents had, the more likely they were to perceive media bias (Lee, 2005).
However, the strongest predictor of perceived media bias in the Lee (2005) study was political cynicism. In a later study including data from more than 47,000 US citizens, Lee (2010) looked at factors influencing another part of credibility – media trust. Here, he again found political cynicism and strong partisanships to be good predictors of trust in media. Because of this, political cynicism might show to have an impact in this study.
However, not only political cynicism but also media cynicism has shown to have an effect on credibility measurements. Using telephone surveys and internet surveys, Tsfati (2010) measured among other things how media scepticism relates to trust in both
mainstream and nonmainstream media and found a connection between the two.
Confirmation bias as well as strong partisanships, political cynicism and media cynicism might influence this study as moderators. Depending on whether the content and the political parties are aligned with the views of the study participants, they might evaluate that content as more credible. If they have strong partisanships, they might evaluate everything as less credible, and depending on the amount of both media and political
cynicism a person conveys, their credibility evaluation might be lower as well. That leads me to the third, fourth, fifth and sixth hypotheses:
H3: People find party press content more credible if they sympathize with the political party behind the content.
H4: The more extreme partisanship in any direction people show, the less credible they will find both traditional media and party press.
H5: The more media cynicism people express, the more likely they are to evaluate both traditional media and party press as less credible.
H6: The more political cynicism people express, the more likely they are to evaluate both traditional media and party press as less credible.
Method
To find out if party press is recognized and evaluated differently than
independent media in terms of credibility, an online experiment was carried out on a sample of Danish citizens. The experiment specifically set out to test how two party press outlets from the two biggest parties in parliament were evaluated in terms of credibility. Dit Overblik (affiliated with right-wing The Danish People’s Party) and Netavisen Pio (affiliated with left-wing The Social Democrats) were chosen. The Danish national broadcasting corporation (DR) was included as a neutral control.
Study design
In a single-factor, pre-tested, between-subjects survey experimental design, participants were randomly assigned to one of the three conditions. A generic issue (from a neutral source) about the government’s intention to close regional councils was selected and manipulated to make it look like it was published on three different media outlet websites. This was done to ensure that effects would be comparable across the three conditions. External validity here was high, because the article was real, and all three outlets had
published articles about the topic earlier. The dependent variable was evaluation of credibility of the content.
Sample
Participants were selected through non-probability convenience sampling between April 8th and April 21st 2019. The survey was sent out through social media and e-mail to reach possible participants. A total of 229 respondents were recruited, but 68 were
later excluded due to not completing the full survey – leaving 161 respondents for the data analysis. 68% were females, 37% were between 25 and 34 of the six age groups [M = 2.83,
SD = 1.50] and 38% had a medium long education (2 to 4,5 years) and 19% had a long
education (of 5 years or longer).
Procedure
The experimental procedure was as follows: Respondents answered a set of demographic questions and were then assigned and exposed to one of the three conditions. The post-test questionnaire contained a manipulation check and the dependent variables. In the analysis after, a randomization check was carried out. On the measures of gender (p = .418), age (p = .958), interest in voting in the two parties tested (p = .167 for The Social Democrats and p = .232 for The Danish People’s Party), and education (p = .069), no significant between groups differences were found.
Stimulus material
Three conditions were used: Right-wing party press, left-wing party press and neutral. The news articles were manipulated versions of an article from DR about the government’s proposal to close the regional councils. Everything was held constant in the three conditions, except: Media outlet specific factors (website header, font and set-up), person quoted (an official (neutral condition) and party leaders of each of the two parties respectively), and the position taken by the parties (visible in quotes, sub-headline and headline). Great effort was put into the layout of the articles in order to make them look as close to the originals as possible (see appendix 2, 3 and 4).
DR was chosen as the neutral, traditional media because it is the most trusted media brand in Denmark according to the latest Reuters Institute Digital News Report
(Schrøder, Blach-Ørsten & Eberholst, 2018). The two other conditions, Netavisen Pio and Dit Overblik, were chosen because they are affiliated with the two biggest parties in the Danish parliament (at the time of the data collection), and because they belong to the right and left respectively. Another interesting difference between them is that they have hidden or highlighted their connection to the parties quite differently which might influence how easy it is for participants to notice the affiliation. The left-wing, Netavisen Pio, does not want to be associated with the party whereas the right-wing, Dit Overblik, speaks openly about the connections (Borre & Rohde, 2018a; Nielsen, 2018).
Manipulation checks
Three manipulation checks were carried out: First, respondents were asked to write the name of the outlet that had published the article in an open-ended question. 18% of the respondents were able to remember the correct outlet name. More named the neutral outlet correctly (N = 15 out of 50) than the left-wing party press (N = 6 out of 60) and the right-wing party press (N =8 out of 51), and the difference between the neutral and the party press outlets is significant, F (1, 16) = 7.29, p = .008. However, all the numbers were very low.
In another manipulation check respondents were asked to fill in the name of the person quoted in the articles. Here, a total of 63% remembered the name of the person quoted in the articles, revealing a lot higher recognition of source names than of outlet name. Of those who saw left-wing party press 43 of 60 named the right person compared to 37 out of 51 for the right-wing party press and 21 out of 50 for the neutral stimuli. The difference between the neutral stimuli and the party press outlets was significant, F (1, 16) = 14.36, p < .001. However, the source in the neutral condition was a relatively unknown official from a nursing organisation compared to the two other stimuli’s well-known politicians.
The last manipulation check was a multiple choice question about what characterises the outlet from where the article came (they had eight options, e.g. “it is
affiliated to a political party”, “it is the national broadcasting corporation”, “it is a regional radio station” “it is affiliated with a charity”). Here 25% recognized the outlet as party press or the national broadcasting corporation, but a further elaboration will be presented in the results section because this manipulation check is also part of the sub-research question. In conclusion, the manipulation checks were successful enough to continue with analyses.
Measures
In order to measure the dependent variable, credibility, respondents were asked to evaluate the articles on a series of statements. Using a version of the questions created by Wang and Hickerson (2016) on the basis of Gaziano and McGrath (1986) they were asked to rate on a 7-point scale from “agree a lot” to “disagree a lot” whether the article was
newsworthy [M = 2.97, SD = 1.38], fair [M = 3.78, SD = 1.27], not biased [M = 4.83, SD = 1.39], telling the entire truth [M = 4.3, SD = 1.43], relevant [M = 2.8, SD = 1.36] and
trustworthy [M = 3.53, SD = 1.4]. Interestingly, these six measures turned out to be extracted as two factors in principal axis factoring – two components had an eigenvalue above 1 (eigenvalue 2.34 and 1.26), but the point of inflection in the scree plot was clear. These factors explain 38.94% and 20.97% of the variance in the initial variables. A direct oblimin
rotation shows that in both the extracted factors all other factors correlate positively with the first. One factor consists of “newsworthy” and “relevant” with “newsworthy” having the strongest association (factor loading .84) and the other factor consists of “not biased”, “fair”, “told the entire truth” and “trustworthy” with “not biased” having the strongest association (factor loading .67). Because of this, the two factors were separated into the “newsworthy” factor [M = 2.89, SD = 1.25] and the “bias” factor [M = 4.11, SD = .93], using the latter as the main dependent variable. Reliability of the newsworthy factor is good with Cronbach’s alpha = .80, and the second factor, the bias factor, is low, but reasonable with Cronbach’s alpha = .60.
For independent variables, a set of demographic questions were asked (see appendix 1). A “propensity to vote” measurement was used to determine participants’
political stance (Blais & St-Vincent, 2010). On a 1-10 scale from “very unlikely” to “very likely” participants were asked “If you could choose between the following parties, how likely is it that you would vote for:” including all parties in government at the time the study was conducted as options. The Social Democrats had the highest score [M = 4.17, SD = 3.35], and The Danish People’s Party had the lowest score of all [M = 1.45, SD = 2.57].
To detect political and media cynicism, respondents were asked, on a 7-point scale from “agree a lot” to “disagree a lot”, to evaluate three statements for political and media cynicism respectively: “Politicians usually tell the truth” [M = 5.30, SD = 1.47], “If we do not keep an eye on politicians they will act solely according to their own interests” [M = 4.86,
SD = 1.53], “Generally speaking, politicians can be trusted” [M = 4.31, SD = 1.49]. These
factored together as one factor and had a low, but reasonable Cronbach’s alpha of .65 [M = 4.82, SD = 1.15]. To determine media cynicism these questions were asked: “You can usually trust what is in the media” [M = 4.16, SD = 1.54], “Media often twist the truth to make a story more interesting” [M = 5.52, SD = 1.23] and “The media often favours one side” [M = 5.12, SD = 1.36]. These also factored together as one factor and reliability was low, but reasonable with a Cronbach’s alpha of .66 [M = 4.93, SD = 1.07]. For all independent variables, see appendix 1 for full transcript of the survey.
Analysis
To test for differences in mean scores of credibility between party press and traditional media, t-tests were conducted. For tests with a moderator – like whether noticing that it was party press or not and how that influences credibility evaluation of the three stimuli - two-way ANOVA analyses were used. The same test was used for moderation in the case
of propensity to vote for certain parties, cynicism and partisanship and their effect on credibility evaluations. Statistical analyses were made using the SPSS software.
Results
In this study, I tested if Danish party press outlets were evaluated differently in terms of credibility compared to traditional media. The main assumption was that party press would be evaluated as less credible – but that many would not recognize it as party press. To test the effects of each of the three conditions in terms of credibility, respondents were asked to rate the article on six credibility items that worked together as two factors, newsworthiness and perceived bias.
Looking at newsworthiness (the newsworthy and relevant items) in a one-way analysis of variance, respondents who saw the neutral stimuli evaluated the article as being quite newsworthy (M = 2.71, SD = 1.31) on the 1 to 7-scale (from agree a lot to disagree a lot). The articles were evaluated as less newsworthy than the neutral by both those who saw the left-wing party press (M = 2.89, SD = 1.17) and the right-wing party press (M = 3.05, SD
= 1.29). A t-test showed that the evaluations from those who saw traditional media were not
significantly different from those who saw party press1, F (2, 19) = .42, p = .234.
However, when looking at the other factor, if the article was perceived to be biased (the not biased, fair, told the entire truth and trustworthy items), results of the one-way
analysis of variance were different. The evaluations from those who saw the neutral stimuli leaned towards a negative evaluation, i.e. the article being slightly biased (M = 4.00, SD = .78). Those who saw left-wing party press evaluated the articles as being less biased (M = 3.94, SD = .85), and the respondents who saw the right-wing party press stimuli evaluated the article as more biased (M = 4.43, SD = 1.07). Comparing those who saw the neutral stimuli with those who saw party press in a t-test did not give a significant result with F (2, 11) = 4.05, p = .2642. For this test, the assumption of equal variances in the population was violated with Levene’s F = (2, 11) = 4.05, p = .046 and the results were interpreted in accordance with this. When splitting the party press outlets into two and comparing them independently with the neutral, one significant result was found. Conducting a t-test, the evaluation of perceived bias in the article showed to be significantly higher than the neutral
1 When conducting t-tests, the two party press outlets were combined and measured together against the neutral. 2 Doing the test with all six credibility items (i.e. without splitting it into ”newsworthy” and ”bias”) gives a similar, not
among those who saw right-wing party press, Mdifference= .43, F (2, 99) = 5.60, p = .0243. Again, the assumption of equal variances in the population was violated with Levene’s F = (2, 99) = 5.60, p = .020 and the results were interpreted in accordance with this. Even though right-wing party press showed to be perceived as more biased than the others, the party press overall were not evaluated as being less credible than traditional media, and therefore hypothesis 1 was rejected.
Table 1
Credibility evaluation
N M SD
Exposure
Right-wing party press Left-wing party press Traditional media 50 51 60 4.43 3.94 4.00 1.07 .85 .78
Note. Credibility evaluation is of the bias factor, not including the newsworthy factors.
Respondents were asked to choose the type of outlet they were presented with on a list of eight items (see appendix 1). Here, more were able to choose the right type for the neutral stimuli (“the national broadcasting corporation” were chosen by 18 out of 50) than the other two (“it is affiliated with a political party” was chosen by 22 out of 111). Even though only less than half of the respondents in the neutral group were able to name the right type of outlet, those who saw party press material did significantly worse, F (1, 16) = 4.92, p = .028. This indicates a negative answer for the sub-research question of whether people are able to recognize party press for what it is. Of course, some were able to recognize that the party press outlets were affiliated with parties, but less than 1/5 did it correctly.
Interestingly, recognizing party press or traditional media for what it is,
significantly, but weakly moderated the evaluation of how biased the stimuli were, r = .14, p = .001 4. Those who recognized the type of outlet they were presented with evaluated the neutral stimuli as more credible (M = 3.72, SD = .99, N = 18 out of 50) and the party press as less credible, especially the right-wing party press (M = 5.18, SD = 1.07, N = 10 out of 51) compared to the left-wing party press (M = 4.4, SD = .93, N = 12 out of 60). Therefore, hypothesis 2, that recognition influences credibility evaluation, can be confirmed.
3 The same test was made with the left-wing stimuli but no significant difference from the neutral was found,
Mdifference= .063, F (2, 11) = .88, p = .691)
*4 As a robustness check, the analysis was also done with both credibility factors (both “newsworthy” and “bias”) and
Table 2
Party press recognition and credibility evaluation
SS df MS F p
Stimuli
Party press recognition Stimuli * party press recognition Error Total 5.50 .624 9.49 124.93 2859.5 1 1 1 157 161 5.50 .624 9.494 .796 6.91 .784 11.93 .009 .377 .001
Note. Dependent variable is the bias factor. N = 161. “Stimuli” is coded as neutral or party press (1 and 0).
Party affiliation turned out to have no statistically significant influence on the evaluation of how biased the stimuli were. A Two-way ANOVA univariate analysis of variance was made by creating a dummy variable for each of the two parties whose party press
outlets were represented. These were made by splitting the propensity to vote scales in below and above the mean score. For The Social Democrats (M = 4.17) those with 5 and above were included and for The Danish People’s Party (M = 1.45) those with 2 and above were included5. In the case of The Social Democrats, support for the party showed no difference in evaluation of the left-wing party press article’s bias, Mdifference= .04, F (1, 16) = .018, p = .892. Supporting The Danish People’s Party did not affect the evaluation of bias in the right-wing party press article either, Mdifference= .63, F (1, 16) = 3.86, p = .0516. Because there was no link between party affiliation and credibility evaluation, hypothesis 3 was rejected.
Very partisan political opinions (those who chose “10” for any party on the 1-10 scale of how likely they were to vote for a party) did not have a significant interaction effect with stimuli on credibility measures, F (1, 16) = .67, p = .512. However, 120 of 161
participants chose “10” for at least one party, and it did therefore not necessarily show to be
5 The decision to split here arguably means that respondents that are less sure that they want to vote for The Danish
People’s Party than of The Social Democrats are included. However, since The Social Democrats is a center-left party compared to the far-right Danish People’s Party it arguably means more if you are even likely (in opposition to sure that you will) vote for the right-wing party.
6 The same analysis made with all credibility measures (both the ”newsworthy” and ”bias” factors) showed a
significant effect on credibility evaluation on the right-wing party press by those who supports The Danish People’s Party. When calculated this way, party supporters evaluate the article as being more credible, Mdifference= .55, F (1, 16)
the best measure of partisanship. Because of this, hypothesis 4 of extremely partisan views influencing the credibility evaluation was rejected.
Possible effects of both political and media cynicism were analysed using a one-way ANOVA with dummy variables of those who showed cynicism towards the media and politicians respectively. Respondents were split in above and below mean score – for media cynicism M = 4.93 and for political cynicism M = 4.83. Political cynicism did not have an effect on credibility evaluation (Mdifference= .16, F (1, 16) = 1.2, p = .275) and neither did media cynicism (Mdifference= .01, F (1, 16) = .003, p = .959). This means that hypotheses 5 and 6 about media and political cynicism influencing credibility evaluation negatively were rejected.
Discussion
Within the past few years, political parties in Denmark have created their own media outlets. This trend has been described by traditional media, but no research has been done on what has been named the re-emerged party press (Jørgensen, 2017; Madsen, 2018; Villumsen, 2019). The purpose of this survey-experiment study was to do an initial test of if – and how – political party press outlets are evaluated differently in terms of credibility compared to traditional media. Furthermore, a sub-research question asked whether people would be able to recognize party press for what it is.
The results showed that respondents did not rate the credibility of party press and traditional news media differently from each other. However, respondents did evaluate the right-wing party press outlet as being significantly more biased than the left-wing party press and traditional media. According to literature, credibility evaluations of news content are highly linked to the perceived credibility of the news media (Cheng & Lo, 2012; Roberts 2010). Furthermore, media outlets that a person knows well and use often are also more often perceived to be credible (Johnson & Kaye, 2000; Kiousis, 2006; Trench & Quinn, 2003). In general, people were not able to identify the party press’ political affiliations. However, the results interestingly showed that those who did recognized the party press as such rated the credibility of the articles significantly lower than those who did not.
What this result indicates is how important it is to be able to recognise party press for what it is in order to be able to differentiate it from traditional media. This brings up the theory of the segmented press, where news media target certain, specific groups of society with their content in order to stay relevant (Schultz, 2007). The results of this study indicate that political parties are succeeding in using this tendency to their advantage, creating their
own media to not just provide their supporters with political views but also news to create a stronger community around themselves.
In this study, neither media cynicism, political cynicism nor strong partisanship showed to be significant predictors of lower perceived credibility. This is in contrast to literature that finds that political cynicism, media cynicism and strong partisanship are all good predictors of media bias perception (Giner-Sorolla & Chaiken, 1994; Lee, 2005; 2010; Tsfati, 2010). One reason for this might be that the sample was too small for the effects to show themselves. It is also possible that the articles were too orderly and factual for both political and media cynicism’s effect on media bias perception to show up. For strong
partisanship, a possible explanation is that the measure did not turn out to be good, because almost all (120 out of 161) respondents fell into the group of strong partisanship. This
possibly happened because the experiment was conducted shortly before a national election and people had already decided who they wanted to vote for.
Furthermore, no relationship between party identification and credibility ratings of the party press content was found, even though confirmation bias has shown to have the effect that people are more likely to find credible what they already know and support (Kim, 2019; Klayman, 1995). A possible explanation is that there were too few in the sample who expressed intention to vote for especially The Danish People’s Party, whereas almost everyone expressed intention to vote for The Social Democrats. However, like literature on online media suggests, identifying the source can be more difficult with online media compared to traditional media, and knowledge of a media company increases recognition (Cheng & Lo, 2012; Freeman & Spyridakis, 2004; Johnson & Kaye, 2000; Rimmer & Weaver, 1987). No statistics on numbers of readers have been published, but because party press in a Danish context is still very new, and because it is arguably more of a niche than the public service corporation it is being compared to, it is fair to suggest that it is less well-known. That might be one of the reasons why there were no differences in the overall credibility
evaluations of party press versus traditional media – that people simply did not recognise the difference between the stimuli. This is further backed by the fact that those who did
recognize the difference rated the credibility differently.
Interestingly, the six credibility factors did not end up correlating with each other as expected. Instead, factor analysis uncovered that they measured two different things: Newsworthiness and bias respectively. This is different from the studies these six items were derived from (Gaziano & McGrath, 1986; Meyer, 1988; Wang & Hickerson, 2016). However, the credibility measure was used differently in this study. Instead of rating the credibility of different articles, the same article (except for small differences) was evaluated on different
platforms. Because of this, this study has methodological implications. The “newsworthy” and “relevant” items showed to be less important factors in measuring credibility across
platforms, whereas “biased”, “told the entire truth”, “trustworthy” and “fair” were more
accurate. When the same article with only small changes is being used, this arguably makes it difficult to dispute its newsworthiness or say that it is more or less relevant.
However, despite the study’s contributions to theory and methods, it of course also has a series of limitations. First of all, internal validity is possibly low because
respondents were viewing the article in an artificial setting. In reality, they might have a bigger possibility of noticing the party press element because they see more than just a photo of a website. Especially if they find their way to the website themselves. Re-running the same study in a more true-to-life setting would be interesting. However, because of social media it is arguably relatively true to real conditions to be asked to quickly identify a source today, which increases the external validity.
Furthermore, the stimuli were kept very simple and as similar as possible in order to be able to compare them with each other afterwards. However, this similarity is a possible explanation for not finding a difference between party press and traditional media. Elaborating on this, the stimuli article was about a very serious political topic – the possible closing of regional councils. Had it instead been an article with a more personal outlook – for instance about one politician criticising another politician – the party press element might have shown to stand out more, and future studies should take this into consideration.
Another reason for the difference between the two party press stimuli could be that it is not equally easy to notice that it is party press. As mentioned, the right-wing party has been very explicit about its ties to the party press website (Nielsen, 2018). This is also visible on the website, for instance because the website has an ad for the party in the header next to the media’s name, and the font and colours of the media’s logo is kept the same as the political party’s logo. The left-wing party, however, does not want to be directly
associated with its party press (Andersen, 2018) and its party press was evaluated as being more credible. This media’s logo and colours are nothing like the logo of the political party (see appendix 2 and 3).
The fact that some political parties are open about their affiliation with their media outlets whereas others are not put a great responsibility on the traditional media to report on the rise of party press, and on the citizens for being able to recognize it. Even when party press claims to be independent from political interests, it is different than traditional media: The most important thing to party press outlets might not be to select and present
news in a fair, unbiased way, but to promote certain views on the world. And in order to be able to navigate through this, citizens need to be able to recognize it for what it is.
Leading up to this study, an investigation of political parties’ reasons for creating party press outlets was conducted as well as political journalists’ possible concerns about it (Jensen, 2018). This was done through qualitative interviews with two political parties and two political journalists and through a small quantitative content analysis of party press content. A more thorough methodological approach, and a bigger sample size, could provide more answers and insight into the reasons for the phenomenon and its possible
consequences for society. A qualitative study with in-depth interviews or focus groups with citizens could shed light on how the re-emergence of party press is perceived and used in the population.
However, the difference between the outlets could also have to do with the fact that the two outlets are not equally credible. Maybe one of the party press outlets resemble traditional media more, and the respondents evaluate the credibility accordingly. In order to know if party press is in fact less credible than traditional media, further studies should be conducted. This e.g. includes quantitative content analyses of sources, topics, the news they decide to cover and how they spin it. A qualitative content analysis of e.g. tone or a critical discourse analysis of discourses in party press compared to traditional media could also greatly improve the theory on the subject.
Another limitation is the respondent group. First of all, only 161 people took part in the survey, with between 50 and 60 in each group. Because of this, some nuances or smaller tendencies might have been lost, and that might be one reason for the lack of support for some of the hypotheses. Moreover, the respondent group was arguably not representative of the general population, with many being young, highly educated, females, left-wing voters. Future studies of credibility perceptions of party press should therefore include more respondents if possible.
Furthermore, within two months after conducting this study, a national election was to take place in Denmark. It is also possible that this influenced the results. The
bandwagon effect gives support to those who are already leading “the race” (Kiss &
Simonovits, 2013). In the case of the Danish election, The Social Democrats were expected to win the election, form government, and their leader was expected to become the new prime minister. As she was the one quoted in one of the articles, a bandwagon effect possibly effected the credibility evaluation of the left-wing party press positively.
This study is a very initial first glance at the re-emergence of party press in Denmark. Following the rise of a segmented media landscape, political parties are now
creating their own media, “infiltrating” what used to be a “clean” media environment. In times of falling support and less party identification, political parties can utilise these new news media platforms to attract the segments of the public that has similar ideas about society. Because of mediatization and years of adapting to media logic, news media arguably taught the political parties how to do well in the media – and now they might as well do it
themselves (Jensen, 2018). Because of social media and the technological advancements, it is easier than ever before – both for an outsider to use the toolbox of journalists but also to hide true intentions online. This study showed the importance of being able to recognise the source of online news in order to evaluate it accordingly.
References
Allern, S. (2007). From party press to independent observers? An analysis of election campaign coverage prior to the general elections of 1981 and 2005 in two Norwegian
newspapers. Nordicom Review, 28(Jubilee issue), 63-79. Retrieved from:
http://www.nordicom.gu.se/sites/default/files/kapitel-pdf/248_248_allern1.pdf
Allern, S. & Blach-Ørsten, M. (2011). The news media as a political institution.
Journalism Studies, 12(1), 92-105. DOI: 10.1080/1461670X.2010.511958
Alsaad, A., Taamneh, A. & Al-Jedaiah, M. N. (2018). Does social media increase racist behaviour? An examination of confirmation bias theory. Technology in Society, (55), 41-46. DOI: 10.1016/j.techsoc.2018.06.002
Alternativet (n.d.). Altivisten. Altivisten.dk. Retrieved from: https://altivisten.dk/ Altheide, D. L. & Snow, R. P. (1979). Media Logic. California: Sage.
Andersen, L. S. (2018, June 2). Chefredaktør for den erklæret uafhængige netavis Pio fik honorarer af Socialdemokratiet. Information. Available from:
https://www.information.dk/indland/2018/06/chefredaktoer-erklaeret-uafhaengige-netavis-pio-fik-honorarer-socialdemokratiet
Andersen, L. S. & Geist, A. (2018, June 1). Den omstridte netavis Pio har deltaget I interne S-møder – socialdemokraterne mørklægger omstændighederne. Information. Available from: https://www.information.dk/indland/2018/05/omstridte-netavis-pio-deltaget-interne-s-moeder-socialdemokraterne-moerklaegger-omstaendighederne
Blais, A. & St-Vincent, S. L. (2010). Personality traits, political attitudes and the propensity to vote. European Journal of Political Research, 50(3), 395-417. DOI: 10.1111/j.1475-6765.2010.01935.x
Blumler, J. G. (2018). The crisis of public communication, 1995-2017. Journal of the
European Institute for Communication and Culture, 25(1-2), 83-92. DOI:
10.1080/13183222.2018.1418799
Blumler, J. G. & Kavanagh, D. (1999). The third age of political communication: Influences and features. Political Communication, 16(3), 209-230. DOI:
10.1080/105846099198596
Borre, M. & Rohde, T. S. (2018a, November 3). Chefredaktør for Pio afviser kritik: Hos os er det en styrke at have arbejdet for et politisk parti. Berlingske. Retrieved from:
https://www.berlingske.dk/politik/chefredaktoer-for-pio-afviser-kritik-hos-os-er-det-en-styrke-at-have
Borre, M. & Rohde, T. S. (2018b, November 3). Kulegravning af omdiskuteret netavis afslører tætte band til Socialdemokratiet. Berlingske. Available from:
https://www.berlingske.dk/politik/kulegravning-af-omdiskuteret-netavis-afsloerer-taette-baand-til
Cheng, B. K. L. & Lo, W. H. (2012). Can news be imaginative? An experiment testing the perceived credibility of melodramatic animated news, news organisations, media use, and media dependency. Electronic News, 6(3), 131-150. DOI: 10.1177/1931243112456531
Couldry, N. & Hepp, A. (2013). Conceptualizing mediatization: Contexts, traditions, arguments. Communication Theory, 23(3), 191-202. DOI: 10.1111/comt.12019
Dalton, R. J., Beck, P. A. & Huckfeldt, R. (1998). Partisan cues and the media: Information flows in the 1992 presidential election. American Political Science Reviews,
92(1), 111-126. DOI: 10.2307/2585932
Dansk Folkeparti (n.d.). Dit Overblik. Ditoverblik.dk. Retrieved from:
https://ditoverblik.dk/
Đorđević, J. (2019). Credibility in the Serbian Media: Argumentation as an Instrument of Trust? In J. Schmied & J. Dheskali (Eds.), Credibility, Honesty, Ethics & Politeness in
Academic and Journalistic Writing (pp. 43-59). Göttingen: Cuvillier Verlag.
Engelke, K. M., Hase, V. & Wintterlin, F. (2019). On measuring trust and distrust in journalism: Reflection of the status quo and suggestions for the road ahead. Journal of Trust
Research, 9(1), 66-86. DOI: 10.1080/21515581.2019.1588741
Enhedslisten (n.d.). Listen – Enhedslistens podcast. Enhedslisten.dk. Retrieved from:
https://enhedslisten.dk/temaer/listen-enhedslistens-podcast
Flanagin, A. & Metzger, M. (2000). Perceptions of Internet information credibility.
Journalism & Mass Communication Quaterly, 77(3), 515-540. DOI:
Fletcher, R. & Park, S. (2017). The impact of trust in the news media on online news consumption and participation. Digital Journalism, 5(10), 1281-1299. DOI:
10.1080/21670811.2017.1279979
Freeman, K. S. & Spyridakis, J. H. (2004). An examination of factors that affect the credibility of online health information. Technical Communication, 51(2), 239-263. Retrieved from:
https://www.researchgate.net/publication/233542280_An_Examination_of_Factors_That_Aff ect_the_Credibility_of_Online_Health_Information
Gaziano, C. & McGrath, K. (1986). Measuring the concept of credibility. Journalism
and Mass Communication Quaterly, 63(3), 451-462. DOI: 10.1177/107769908606300301
Giner-Sorolla, R. & Chaiken, S. (1994). The causes of hostile media judgments.
Journal of Experimental Social Psychology, 30(2), 165-180. DOI: 10.1006/jesp.1994.1008
Golan, G. J. (2010). New perspectives on media credibility research. American
Behavioural Scientist, 54(1), 3-7. DOI: 10.1177/0002764210376307
Gregersen, R. V. (2018). Nyhedsmediernes historie. In Medier NU. Systime: Aarhus. Retrieved from: https://mediernu.systime.dk/index.php?id=frontpage&cmd=toc
Hepp, A., Hjarvard, S. & Lundby, K. (2015). Mediatization: Theorizing the interplay between media, culture and society. Media, Culture and Society, 2(37), 314-324. DOI: 10.1177/0163443715573835
Hjarvard, S. (1995). Nyhedsmediernes rolle i det politiske demokrati. Statsministeriets medieudvalg: Denmark. Retrieved from:
https://www.researchgate.net/publication/288822404_Nyhedsmediernes_rolle_i_det_politisk e_demokrati
Hjarvard, S. (2007). The political press. An analysis of political parallelism in Danish newspapers. Journalistica, 2(5), 27-53. Retrieved from: http://newspaperinnovation.com/wp-content/uploads/stighjarvard.pdf
Hjarvard, S. (2008). The mediatization of society: A theory of the media as agents of social and cultural change. Nordicom Review, 29(2), DOI: 10.1515/nor-2017-0181
Høyer, S. (2005). The Rise and Fall of the Scandinavian Party Press. In S. Høyer & H. Pöttker (Eds.), Diffusion of the News Paradigm 1850-2000 (pp. 75-92). Gothenburg: Nordicom.
Jensen, S. (2018). The re-emergence of ‘party press’ in Denmark: Why is it
happening and what does it mean for journalists? (Exam for Media and Journalism Research in the Context of Globalization). University of Aarhus. Available from:
https://www.researchgate.net/publication/333247784_The_Re-emergence_of_'Party_Press'_in_Denmark_Why_is_it_happening_and_what_does_it_mean_ for_journalists
Johnson, T. J. & Kaye, B. K. (2000). Using is believing: The influence of reliance on the credibility of online political information among politically interested internet users.
Journalism & Mass Communication Quaterly, 77(4), 865-879. DOI:
10.1177/107769900007700409
Jørgensen, P. S. (2017, March 19). Partipressen er på vej tilbage – og det er ok for demokratiet. Berlingske. Retrieved from: https://www.berlingske.dk/kronikker/partipressen-er-paa-vej-tilbage-og-det-er-ok-for-demokratiet
Kalogeropoulos, A. (2018). Which brands do we trust and why? In Newman, N. (Ed.)
Reuters Institute Digital News Report 2018 (pp. 42-44). Reuters Institute for the Study of
Journalism. Retrieved from: http://media.digitalnewsreport.org/wp-content/uploads/2018/06/digital-news-report-2018.pdf
Kim, K. (2019). The hostile media phenomenon: Testing the effect of news framing on perceptions of media bias. Communication Research, 36(1), 35-44. DOI:
10.1080/08824096.2018.1555659
Kim, M. (2014). Partisans and controversial news online: Comparing perceptions of bias and credibility in news content from blogs and mainstream media. Mass Communication
and Society, 18(1), 17-36. DOI: 10.1080/15205436.2013.877486
Kiousis, S. (2006). Exploring the impact of modality on perceptions of credibility for online news stories. Journalism Studies, 7(2), 348-359. DOI: 10.1080/14616700500533668
Kiss, Á. & Simonovits, G. (2013). Identifying the bandwagon effect in two-round elections. Public Choice, 160(3), 327-344. DOI: 10.1007/s11127-013-0146-y
Klayman, J. (1995). Varieties of confirmation bias. Psychology of Learning and
Motivation, 32, 385-418. DOI: 10.1016/S0079-7421(08)60315-1
Langberg, M. (2018, March 2). Alternativet i aktiv protest mod konflikt-journalistik.
Journalisten.dk. Retrieved from https://journalisten.dk/alternativet-i-aktiv-protest-mod-konflikt-journalistik
Lee, T. (2005). The liberal media myth revisited: An examination of factors influencing perceptions of media bias. Journal of Broadcasting & Electronic Media, 49(1), 43-64. DOI: 10.1207/s15506878jobem4901_4
Lee, T. (2010). Why they don’t trust the media: An examination of factors predicting trust. American Behavioural Scientist, 54(1), 8-21. DOI: 10.1177/0002764210376308
Lessel, S. (2018, August 8). Chefredaktør for nyt DF-medie: Partiet dikterer ikke indholdet. Men … Altinget.dk. Retrieved from: https://www.altinget.dk/artikel/chefredaktoer-for-nyt-df-medie-df-bestemmer-ikke-hvad-jeg-maa-skrive
Madsen, M. (2018, August 10). Partipressen er tilbage: “Nu kan partierne selv styre de budskaber, de ønsker at få frem i offentligheden”. Information. Retrieved from:
https://www.information.dk/kultur/2018/08/partipressen-tilbage-kan-partierne-styre-budskaber-oensker-faa-frem-offentligheden
McGrew, S., Breakstone, J., Ortega, T., Smith, M. & Wineburg, S. (2018). Can students evaluate online sources? Learning form assessments of civic online reasoning.
Theory & Research in Social Education, 46(2), 165-193. DOI:
10.1080/00933104.2017.1416320
McGrew, S., Ortega, T., Breakstone, J. & Wineburg, S. (2017). The challenge that’s bigger than fake news: Civic reasoning in a social media environment. American Educator,
41(3), 4-9.
Meyer, P. (1988). Defining and measuring credibility of newspapers: Developing an index. Journalism & Mass Communication Quarterly, 65(3), 567–574. DOI:
10.1177/107769908806500301
Meyer, H. K., Marchionni, D. & Thorson, E. (2010). The journalist behind the news: Credibility of straight, collaborative, opinionated, and blogged “news”. American Behavioural
Scientist, 54(2), 100-119. DOI: 10.1177/0002764210376313
Ministry of Culture. (2017). Mediestøtte. Slks.dk. Retrieved from
https://slks.dk/medier/mediestoette/
Netavisen Pio (n.d.a). Netavisen Pio. Piopio.dk. Retrieved from: https://piopio.dk/ Nickerson, R. S. (1997). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), 175-220. DOI: 10.1037/1089-2680.2.2.175
Nielsen, M. (2018, August 7). DF lancerer nyt medie – og det bliver ikke dem, der går hårdest til Messerschmidt, erkender redaktør. TV2.dk. Available from:
http://nyheder.tv2.dk/politik/2018-08-07-df-lancerer-nyt-medie-og-det-bliver-ikke-dem-der-gaar-haardest-til-messerschmidt
Oyedeji, T. A. (2010). The credible brand model: The effects of ideological
congruency and customer-based brand equity on news credibility. American Behavioural
Scientist, 54(2), 83-99. DOI: 10.1177/0002764210376312
Rimmer, T. & Weaver, D. (1987). Different questions, different answers? Media use and media credibility. Journalism & Mass Communication Quarterly, 64(1), 28-44. DOI: 10.1177/107769908706400104
Roberts, C. (2010). Correlations among variables in message and messenger credibility scales. American Behavioural Scientist, 54(1), 43-56. DOI:
10.1177/0002764210376310
Rohde, T. S. & Jessen, C. K. (2018, August 24). Intern mail fra Henrik Sass sender omdiskuteret medie i modvind: ”Det er tydeligvis én stor løgn”. Berlingske. Retrieved from:
https://www.berlingske.dk/politik/intern-mail-fra-henrik-sass-sender-omdiskuteret-medie-i-modvind-det-er-tydeligvis
Schrøder, K. C., Blach-Ørsten, M. & Eberholdt, M. K. (2018). Danskernes brug af nyhedsmedier 2018. Center For Nyhedsformidling. Roskilde: Denmark. DOI:
10.5281/zenodo.1285599
Schultz, I. (2007). Fra partipresse over omnibuspresse til segmentpresse.
Journalistica, 2(5), 5-26. Retrieved from: https://tidsskrift.dk/journalistica/article/view/1807 Strömbäck, J. (2008). Four phases of mediatization: An analysis of the mediatization of politics. The International Journal of Press/Politics, 13(3), 228-246. DOI:
10.1177/1940161208319097
Trench, B. & Quinn, G. (2003). Online news and changing models of journalism. Irish
Communications Review, 9(1), 1-11. DOI: 10.21427/D7TQ65
Tsfati, Y. (2010). Online news exposure and trust in the mainstream media: Exploring possible associations. American Behavioural Scientist, 54(1), 22-42. DOI:
10.1177/0002764210376309
Villumsen, K. V. (2019, March 24). Demokratiforsker: Partipressen er tilbage. Mandag
Morgen. Retrieved from: https://www.mm.dk/artikel/demokratiforsker-partipressen-er-tilbage Wang, X., & Hickerson, A. (2016). The role of presumed influence and emotions on audience evaluation of the credibility of media content and behavioural tendencies. Journal
of Creative Communications, 11(1), 1-16. DOI: 10.1177/0973258616630196
Wayne, W. & Hu, Y. (1994). The effects of credibility, reliance, and exposure on media agenda-setting: A path analysis model. Journalism & Mass Communication Quarterly,
71(1), 90-98. DOI: 10.1177/107769909407100109
Westerwick, A., Kleinman, S. B. & Knobloch-Westerwick, S. (2013). Turn a blind eye if you care: Impacts of attitude consistency, importance, and credibility on seeking of political information and implications for attitudes. Journal of Communication, 63(3), 432-453. DOI: 10.1111/jcom.12028
Appendix 1: Survey experiment (translated from Danish to English) --- initial disclaimer ---
Dear participant, Welcome!
In this survey you will be asked about your opinion on a series of general political questions. You will also be asked some questions about you as a person. They are for the analysis of the results afterwards.
It is important to emphasize that your participation is completely anonymous. It will not be possible to identify each participant, and this means you can answer the questions freely in accordance with your views. The answers will be used for academic purposes. Under no circumstances will anyone from outside the institution have access to your data.
Your participation is completely voluntary, and if you decide you do not wish to participate in the study after all, you can stop at any time. If you already filled out the survey, but would like to withdraw your participation, you have seven days after your participation to do so. If you have concerns about the survey or have questions you can write the ethical committee at: ASCoR Secretariat Ethics Committee University of Amsterdam PO Box 15793 1001NG Amsterdam Tel: +31 20 525 3680 Email: ascor-secr-fmg@uva.nl
The survey will take approximately 10 minutes to complete. Thank you in advance,
Sanni Jensen
Email: sanni.jensen@student.uva.nl
Please check the box below if you want to participate in the study. You cannot go back to previous questions once you have moved forward to the next page.
By checking the box, you agree on the following:
- I want to participate in this study and I understand that my participation is completely anonymous.
- I have read the information given to me about participating in this study. - I am above the age of 18.
Please take your time to read and answer each question, and please be sure to have understood the question before you answer.
Thank you in advance!
--- Personal profile section ---
We will begin with some questions about you as a person. What is your sex?
[ ] Female [ ] Male [ ] Other
How old are you? (drop-down) - 18-24 - 25-34 - 35-44 - 45-54 - 55-64 - 65-74 - 75-84 - 85 or older
What is the highest level of education you have completed?
[ ] Basic school (ex. Public school or realskole) (ß last ex. is an older form of public school, ed.l)
[ ] Vocational training (ex. EUD or workman / craftsman) (ß first ex. is a type of school for practical learning, ed.)
[ ] High school (ex. HF and HTX) (ß these are two types of high schools in Denmark, ed.)
[ ] Short higher education (maximum of two years long)
[ ] Medium long higher education (between 2 and 4.5 years long) [ ] Long higher education (5 years or more)
[ ] Other
--- Opinions and attitudes section ---
On this scale, “0” means that it is very unlikely that you are going to vote for the party while “10” means that it is very likely.
(All options have a slider from 0 to 10, ed.) - Enhedslisten - SF - Alternativet - Socialdemokratiet - Radikale Venstre - Venstre - Liberal Alliance
- Det Konservative Folkeparti - Dansk Folkeparti
To what extend do you agree or disagree with the following statements?
Scale: Agree a lot, agree, somewhat agree, neither or, somewhat disagree, disagree, disagree a lot
Questions/statements:
- Politicians typically tell the entire truth
- If we do not keep an eye on elected officials, they will act in accordance with only their own interests
- Politicians can generally be trusted
To what extend do you agree or disagree with the following statements?
Scale: Agree a lot, agree, somewhat agree, neither or, somewhat disagree, disagree, disagree a lot
Questions/statements:
- You can usually trust what is in the news media
- The news media often bend the trust to make a story more interesting - News media typically favours one part
--- Stimuli ---
On the next page you will see a short article. Please read the entire article before you move on to the next question.
The “next” button will automatically show after 10 seconds. - Party press left (The Social Democrats’ Netavisen Pio)
or
- Party press right (The Danish People’s Party’s Dit Overblik) or
- Traditional media (The Danish Broadcasting Corporation, DR) --- Manipulation check section ---
On which news media was the article you just read published? If you cannot remember, you can write “do not know”.
[ ] (Open answer question, ed.)
What characterises that media outlet? [ ] It is affiliated with a political party [ ] It is a local newspaper
[ ] It is affiliated with a charity organisation [ ] It is a regional radio station
[ ] It is a personal blog
[ ] It is the national public service-company [ ] I did not recognize the media outlet [ ] Don’t know
Who was cited in the article? If you cannot remember, you can write “do not know”. [ ] (Open answer question, ed.)
--- Post-test section ---
Think about the article you just read. Do you agree or disagree that the article:
Scale: Agree a lot, agree, somewhat agree, neither agree nor disagree, somewhat disagree, disagree, disagree a lot
Statements/ questions: - Was newsworthy - Was fair
- Was biased
- Told the entire truth - Was relevant - Was credible
--- End notes --- Important!
The article you read had been constructed for the purpose of this study, and it does not necessarily reflect the truth.
The purpose of the study was to test how different types of media outlets are being perceived as more or less credible.
If you are interested in learning the results after the study had been conducted, you can write an e-mail to sanni.jensen@student.uva.nl.
If you have any comments about the study or the survey you just filled out, you can leave a comment beneath.
Afterwards, please click on the arrow to move on to the last page. [ ] (Open answer question, ed.)
End:
Appendix 2: Left-wing party press stimuli
S om regeringens sundhedsudspil:
Skaber mindre nærhed for patienter
”Det lyder som en rigtig dårlig ide,” siger Mette Frederiksen
om regeringens nye sundhedsudspil, der blandt andet
indebærer at nedlægge regionsrådene.
De folkevalgte regionsråd skal lukke, hvis det står til regeringen.
Thomas Nielsen
og
Signe Lund
16. januar 2019 – 16:33
Landets fem regionsråd må lade livet, hvis vi skal skabe et sundhedsvæsen, der er
tættere på borgerne.
Sådan lød det fra statsminister Lars Løkke Rasmussen (V), da han tidligere i dag
præsenterede regeringens udspil til en sundhedsreform.
Formand for Socialdemokratiet, Mette Frederiksen, giver dog ikke meget for den ide.
”Det løser ikke de problemer, der er i sundhedsvæsenet. Jeg frygter for de
medarbejdere, der skal igennem en masse omstillinger, hvis reformen bliver en
realitet. Det går i sidste ende ud over borgerne,” siger hun.
I den nye struktur, som regeringen lægger op til, skal regionerne bestå som
administrative enheder - men uden det folkevalgte led. Det vil sige, at regeringen vil
nedlægge de fem regionsråd og vinke farvel til de 205 politikere, der i dag
bestemmer i regionerne.
I stedet skal de fem regioner fungere som administrative sundhedsforvaltninger med
embedsfolk.
Herunder har regeringen planer om at etablere 21 sundhedsfællesskaber, der skal
sikre mere nærhed i sundhedsbehandlingen.
”I dag sidder regionsrådet tæt på sygeplejerskerne og lægerne, men hvis det bliver
professionelle bestyrelser, der skal træffe de tunge beslutninger rundt om i landet, så
er jeg da bekymret for, hvordan de lokale får indflydelse,” siger Mette Frederiksen.
Regionerne driver i dag hospitalerne. I hver af de fem regioner i Hovedstaden,
Sjælland, Syddanmark, Midtjylland og Nordjylland sidder der 41
regionsrådspolitikere.
Fire af de fem regioner er ledet af socialdemokrater, mens venstrepolitikeren
Stephanie Lose er formand for Region Syd.
Appendix 3: Right-wing party press stimuli
DF om regeringens sundhedsudspil:
Skaber mere nærhed for patienter
Af Redaktionen – 16. januar 2019De folkevalgte regionsråd skal lukke, hvis det står til regeringen. Foto: Mads Claus Rasmussen
