DEPARTMENT OF JOURNALISM, MEDIA AND COMMUNICATION Agenda-Setting in the 2023 Dutch National Elections Deniz Dönmez Master’s Thesis in Media and Communication Thesis: 30 hp Program and/or course: Master’s Programme in Political Communication/MK2502 Level: Second Cycle Term/year: Spring/2024 Supervisor: Dr. Nicklas Håkansson Examinor: Dr. Isabella Glogger Number of pages: 73 incl. references & appendices Number of words: 20.903 Abstract This thesis examines the interaction between the media agendas of the newspapers De Telegraaf (tabloid) and NRC (broadsheet) and the political agendas of four major political parties, PVV, GroenLinks-PvdA, VVD and NSC, as well as the role of professionalised political campaigning in the interaction between the media and political agendas during the 2023 Dutch national elections. Employing a quantitative content analysis, this study explores the media and political agendas through mentions of issues and political leaders across 487 newspaper articles and 636 Facebook posts and advertisements. Although the interplay between media political agendas has been studied in the past, the Dutch contemporary elections in 2023 have not been researched before, and the role of professionalisation in political campaigns has received less attention as part of the interplay between these agendas. Findings reveal distinct differences between the tabloid and broadsheet media agendas. De Telegraaf emphasised immediate, tangible issues such as taxes and public transport, alongside nature and environment and immigration, reflecting its editorial emphasis on current daily life events. NRC concentrated on long-term societal problems such as nature and environment and education and science. Both newspapers gave considerable coverage to political leaders Pieter Omtzigt and Frans Timmermans, highlighting their significant roles in the political discourse. This study further identified varied degrees of professionalisation in political campaigns. PVV exhibited many indicators of professionalised campaigns and has a significant positive correlation between their political agenda and the media agendas on the issues of nature and environment and immigration. Conversely, GroenLinks-PvdA, despite a highly professional campaign, showed no significant correlations between its political agenda and the media agendas. VVD (high professionalisation) and NSC (low professionalisation) also exhibited no significant correlation between their political agendas and media agendas. This reveals the complex interaction between the media and political agendas and the inconsistent role of professionalisation within this relationship. External factors, such as institutional factors, timing, content characteristics and a difference in campaign tools used may be an explanation for this complexity. By highlighting the interaction between the political and media agendas, this study underscores the need for political communicators, media professionals, and political strategists to understand (digital) media platforms and agenda-setting in the public discourse. Future research should explore these dynamics in different electoral contexts, platforms, and methods to further understand their implications. Keywords: agenda-setting, media agenda, political agenda, Dutch national elections, social media, political campaigns, professionalised campaigns Preface While writing my thesis, I have concluded that understanding and working with campaign strategies, voter behaviour, and agenda-setting motivate me academically and personally. There are many people I would like to thank. First, my supervisor, Dr. Nicklas Håkansson, for his insights and guidance during the research process. Additionally, I want to thank my girlfriend Franka, her father Henk, and my cat Milo for their emotional and academic support. Without you, I would not have been able to write this thesis. Lastly, I want to thank my mother for everything she has done for me. Without her, I would not have been able to come where I am. Sadly, she cannot see my achievements, but she will always be here. Table of Contents 1. Introduction .......................................................................................................................... 1 1.1. Aim of the Study .............................................................................................................. 4 1.2. Research Questions .......................................................................................................... 4 2. Theoretical Background ...................................................................................................... 6 2.1. Agenda-Setting Theory and Media Influence .................................................................. 6 2.2. The Media Agenda ........................................................................................................... 6 2.2.1. Changing Media Landscape ...................................................................................... 7 2.2.2. Differences in Media Agendas: Tabloid vs Broadsheet ............................................ 8 2.3. The Political Agenda........................................................................................................ 9 2.4. Professionalisation of Political Campaigns ................................................................... 10 2.4.1. Indicators of Professionalised Campaigns .............................................................. 12 3. Methods ............................................................................................................................... 15 3.1. Research Design ............................................................................................................ 15 3.2. Data Collection and Processing ..................................................................................... 17 3.2.1. Sampling ................................................................................................................. 17 3.2.2. Data Collection ....................................................................................................... 18 3.2.3. Processing ............................................................................................................... 18 3.3. Corpus ............................................................................................................................ 19 3.4. Codebook ....................................................................................................................... 20 3.5. Reliability of the Data .................................................................................................... 21 3.6. Analysis ......................................................................................................................... 22 4. Results ................................................................................................................................. 23 4.1. Issues and Political Leaders in the Media Agendas ....................................................... 23 4.2. Issues and Political Leaders in the Political Agendas .................................................... 31 4.3. Professionalisation of Political Campaigns ................................................................... 36 4.4. Relationship between Political and Media Agendas ...................................................... 43 5. Conclusion and Discussion ................................................................................................ 49 5.1. Conclusion ..................................................................................................................... 49 5.2. Discussion ...................................................................................................................... 52 References ................................................................................................................................ 54 Appendix 1 – Newspaper Codebook ........................................................................................ 63 Appendix 2 – Facebook Content Codebook ............................................................................ 68 1. Introduction On the 7th of July 2023, the Dutch government collapsed. On this day, public service broadcaster NOS wrote: “Cabinet has fallen, no agreement on asylum measures” (NOS, 2023). After a disagreement on asylum measures and Mark Rutte announcing to quit politics after 17 years of being the Dutch prime minister, the sudden collapse of the Dutch government made many people wonder, what is next? For many political parties, the new elections in November 2023 posed a new political momentum. Significant changes in the political landscape emerged in the run-up to these elections. Notably, a new party, Nieuw Sociaal Contract (NSC), led by former long-time Cristian Democratic Appeal (CDA) Member of Parliament Pieter Omtzigt, was established. The Labour Party (PvdA) and Green Left (GroenLinks) decided to form a joint list led by Frans Timmermans. Other significant parties also saw a leadership change, such as the coalition members People’s Party for Freedom and Democracy (VVD), led by the new leader and Minister of Justice and Safety Dilan Yeşilgöz-Zegerius and Democrats 66 (D66), led by new leader and Minister for Climate and Energy Rob Jetten. Another major party, The Party for Freedom (PVV), had no leadership change and was still led by Geert Wilders. During the final two months of the election campaign, an essential question dominated the news: Which issue(s) will dominate the elections? Will it be immigration, relating to the cabinet's fall, or will it be about social security, the issue primarily associated with Pieter Omtzigt? In September 2023, broadcaster EenVandaag described social security as the voters' most important issue (de Jong & Theulings, 2023). In November 2023, the polling organisation Ipsos published that next to social security, housing, healthcare, immigration, climate, and sustainability are the most important issues for voters (Ipsos, 2023, p. 6). Now, six months after the elections unfolded, the political terrain has shifted. The PVV emerged as the biggest party with 23,5% of the votes and 37 seats in parliament (Kiesraad, 2023). The day after the elections, the newspaper NRC had the headline: “An unprecedented victory for the PVV – and what now?” and De Telegraaf being: “The Netherlands turns right” (Schyns, 2023). The second party of the Netherlands, GroenLinks-PvdA, led by Frans Timmermans, came in at 15,8%, a significant difference from the PVV (Kiesraad, 2023). The VVD came in as the third party with 15,2% of the votes, losing 6,7% of the votes (Kiesraad, 2023). The newcomer NSC came fourth with 12,8% of the votes (Kiesraad, 2023). 1 Dominating the Agendas This significant shift in the Dutch political landscape poses an interesting context for exploring the issues and political leaders that dominated the media and political agendas. Newspapers act as essential bridges between political parties and the public, shaping public perception and impacting the democratic process significantly (Vermeer et al., 2022, p. 8). The difference in coverage, the way they write about certain issues, between tabloid and broadsheet media is relevant to this. For example, Esser & Strömback (2014) suggest that commercial media, like tabloids, portray politics as a game of strategy and focus more on specific issues than others to receive attention (p. 19, 120). Media prioritisation of issues captures public attention and moulds societal perceptions, influencing the democratic process. Vermeer et al. (2022) state: “Agenda setting theories hypothesise that the media agenda drives the public agenda and that media thus determine which political parties or politicians are considered important or relevant by voters.” (p. 3). Data from The Dutch Parliamentary Election Survey (DPES) underscores the significance of the media. Notably, 53,4% of Dutch citizens follow politics on various platforms (radio, TV, etc.) fairly or very closely, and half of the Dutch citizens read the newspaper (almost) daily or a few times a week (Dutch Parliamentary Election Survey, 2017). In contemporary democratic societies, the use of social media as another campaign platform has increased over the last decade, significantly impacting the agenda-setting capabilities of political campaigns (Feezell, 2018, p. 491). Gilardi et al., (2022) describe social media as having “(…) reduced the gatekeeping power of traditional media (…).” (p. 42). The agenda- setting function of social media is relevant to people and in understanding the relationship between the media and political campaigns on social media. Similar to more ‘formal’ political communication (such as press releases), social media gives political campaigns more freedom to express their opinions and have a potentially larger audience (Meiserli, 2021, p. 72). In their study on the Swiss national elections in 2019, Gilardi et al., (2022) acknowledge the importance and complexity of the relationship between social media content by campaigns and the media agenda. Online communication by political campaigns has the potential to reach the agenda of traditional media and thus reach a large audience (p. 54). 2 Professionalisation of Political Campaigns In their analysis of the Dutch media landscape, van der Pas et al. (2017) describe the Dutch media as independent and having high levels of professionalisation (p. 492). This context is crucial for understanding the symbiotic relationship between politicians (political parties) and journalists (media). Politicians gain from having their concerns publicly shared, while journalists gain credibility with their reports and articles, possibly influencing the agenda of members of parliament (van der Pas et al., 2017, p. 506). The interplay between the media agendas and political parties’ agendas has become increasingly significant in the era of more professionalised election campaigns. As Gibson & Römmele (2001) specified many years ago, political campaigns have shifted from conventional, spontaneous, and volunteer-centric approaches to more systematic, strategy- oriented, and expert-driven methods. This evolution, known as professionalisation, signifies a paradigm shift towards sophisticated technologies, data analytics, and intricate communication strategies (p. 35). This strategic approach also potentially affects agenda- setting practices and election outcomes (Meyer et al., 2017, p. 286). Panagopoulos (2006) specified the impact of hiring professional campaign consultants. “(…) the presence of a professional consultant on a campaign increases the candidate’s voteshare by 5 percentage points on average (…)” (p. 867). Political parties now have a more significant influence over the issues given priority in the public discourse thanks to the professionalisation of campaigns. Campaigns can raise awareness of particular issues by using sophisticated data analytics to identify and target specific demographic segments with customised messages (de Ridder et al., 2020, p. 259). This effect is amplified by the deliberate use of social media and digital advertising, which allows parties to directly influence public opinion and conversation without going through traditional media gatekeepers (Scheffauer et al., 2023, p. 2). In addition to encouraging a more involved public, this direct line of communication with voters challenges the agenda-setting function of traditional media. It can change the nature of media influence (Gilardi et al., 2022, p. 42). The contemporary relationship between media and political agendas is very relevant for political communicators, where those who have professional campaigns and are skilled in campaign communication could increasingly hold the power to shape public opinion, advance the democratic discourse and create an informed and engaged citizenry. 3 1.1. Aim of the Study This study aims to explore the interaction between the media agenda, in Dutch newspapers De Telegraaf and NRC and the political agenda of four major political parties, PVV, GroenLinks- PvdA, VVD and NSC, as well as the role of professionalised political campaigning in the interaction between the media and political agendas during the 2023 Dutch national elections. This research seeks to illuminate this relationship by examining how political issues and leaders are highlighted in a contemporary Dutch election. It contributes to a better understanding of the media’s and political parties’ agenda-setting function, and through the viewpoint of campaign professionalisation. 1.2. Research Questions The study's central research question is: What is the relationship between the media agendas of the broadsheet NRC and the tabloid De Telegraaf and the political agendas of PVV, GroenLinks-PvdA, VVD and NSC on Facebook during the final two months of the 2023 Dutch National elections, and what role does the professionalisation of political campaigns have in the interaction between these agendas? Sub-questions: 1. Which issues and political leaders are most prominently featured in the media agendas in the last two months of the 2023 Dutch national elections? a. How do the media agendas differ between broadsheet (NRC) and tabloid (De Telegraaf)? 2. Which issues and political leaders are most prominently featured in the political agendas in the last two months of the 2023 Dutch national elections? 3. What are the professionalisation practises of the political parties, and to what extent do they differ in their degree of professionalisation in the last two months of the 2023 Dutch national elections? 4. To what extent do the political agendas and the media agendas share the same agenda in the last two months of the 2023 Dutch national elections, and what role does professionalisation have in the interaction between these agendas? 4 Study organisation This thesis begins with a theoretical background that addresses agenda-setting theory, media agendas, political agendas, and the professionalisation of political campaigns. The methodology section outlines the quantitative approach used to analyse the datasets included in this research. The results chapter presents how the different newspapers and political parties cover political issues and leaders, what indicators of professionalisation are identified in the political campaigns on Facebook, if the media and political agendas correlate and what role professionalisation plays in this relationship. The thesis concludes with a conclusion and discussion of the main findings, answering the central research question and providing recommendations for further research. 5 2. Theoretical Background 2.1. Agenda-Setting Theory and Media Influence As originally defined by McCombs and Shaw (1972), agenda-setting theory posits that the media significantly influences public perception by determining which issues are deemed important (p. 177). This process of prioritising certain issues over others shapes what the public thinks about and how they think about these issues, a concept known as ”priming” (Weaver, 2007, p. 145). The relevance of the agenda-setting theory becomes particularly evident during election campaigns when media coverage significantly impacts the salience of political issues and candidates (McLaughlin & Velez, 2019, p. 23). For instance, Druckman (2004) discovered evidence of a campaign priming effect in U.S. Senate Elections. He argues that through the media, campaigns can influence the criteria on which voters base their votes by shaping the issues and images displayed (p. 589). The agenda-setting function of the media operates through two primary levels. The first level, known as issue salience, involves the media highlighting specific issues, thereby signalling to the public that these issues are important. The second level, known as attribute salience, concerns how the media focuses on certain aspects of these issues, shaping the public’s perception of the issues’ attributes and the actors involved (McCombs, 2005, p. 546). This distinction helps to understand how media not only tells the public what to think about but also how to think about those issues. 2.2. The Media Agenda The interaction between media and political actors is dynamic, with media that can both lead and follow political agendas depending on the context and the type of issues involved (Barberá, et al., 2019, p. 888). Longitudinal studies which track changes in media influence over time reveal trends and shifts in media's agenda-setting role, suggesting evolving dynamics in how issues are prioritized (McCombs, 2014, p. 117). Research indicates that media can independently impact the political agenda in addition to real-world events (Kleinnijenhuis et al., 2007, p. 378). For instance, media influence tends to be stronger with symbolic issues and foreign policy, while there is more interplay with political agendas on domestic issues (Walgrave & Van Aelst, 2006, p. 93). During election campaigns, the media’s focus on particular issues in election campaigns can shape the public’s perception of candidates and parties. Media coverage can elevate certain 6 issues, thus influencing political agendas (McLaughlin & Velez, 2019, p. 23). For example, during the 2014 European Parliamentary elections, media coverage significantly influenced discourse on EU issues, demonstrating the media’s powerful agenda-setting role (Maier et al., 2017, p. 10). In his analysis, Entman (1993) emphasises the importance of framing in media agenda- setting, suggesting that selecting issues can significantly influence public perception by highlighting certain aspects of an issue while omitting others. This selective emphasis can guide audiences towards a particular understanding or interpretation of events (p. 54). For example, when media outlets repeatedly focus on "economic instability", they can shape a broader view of economic uncertainty, influencing public sentiment and the political agenda. Contemporary research has expanded on this. For instance, a recent study by Feezell (2018) highlights how news outlets on social media have become crucial in shaping the agenda by providing incidental news exposure and social filtering. This process can lead to significant shifts in public opinion as people encounter news within their social networks, which often emphasizes certain issues more than others (p. 490). The emphasis on specific issues is related to media interventionism. Strömback & Dimitrova (2011) specify this by suggesting that “Media interventionism […] refers to the degree to which the content of political news is shaped by journalistic interventions.” (p. 35). For example, journalists have growingly taken on an interpretive reporting style rather than a descriptive one (Strömback & Dimitrova, 2011, p. 35). However, while much research has shown the agenda-setting influence of the media, it is important to note that many external factors can influence the agenda-setting power of the media. Moderators like institutional factors (e.g. type of government), timing and content characteristics (e.g. tone or frames) make that it is not univocal, for example concerning the size of the impact of the media (Vliegenthart & Walgrave, 2019, p. 274). 2.2.1. Changing Media Landscape While traditional media have a significant historically played a significant role in agenda- setting, the media landscape is changing. The ability of traditional news media to influence public opinion, particularly in terms of agenda-setting, may be diminishing. Bennet & Iyengar (2008) describe this as the “era of minimal effects”, where media primarily reinforce existing worldviews rather than altering them (p. 724). Conversely, Esser & Strömback (2014) offer an 7 alternative perspective, highlighting the continued relevance of traditional news media in setting agendas. Different types of media vary in their agenda-setting function and individuals who consume a variety of online news sources may be less susceptible to the agenda-setting influence of traditional media (p. 24, 59). In the Netherlands, the use of traditional news media has declined over the past decade (Commissariaat voor de Media, 2023, p. 4). Despite no longer being the primary news source, traditional media still play a role in conveying political information to citizens. A study on the 2021 Dutch national elections by Vermeer et al. (2022) found that media attention for political parties in traditional outlets still provided electoral benefits (p. 13). The role that digital media play in the agenda-setting relationship between the media and political agendas is very relevant in this digital age. In the formation of the media and political agendas, the concept of a hybrid media system can help us to understand the contemporary dynamic between the agendas. Chadwick (2017) describes the hybrid media system as a system where the rapid reporting capabilities of digital media complement the coverage of traditional media, resulting in a more nuanced portrayal of political issues (p. 82). The immediacy and reach of digital media offer different opportunities for setting the media and political agendas (Harder et al., 2017, p.3). This hybridity creates a more complex and interactive media environment relevant to this research. 2.2.2. Differences in Media Agendas: Tabloid vs Broadsheet The difference between tabloid and broadsheet news media outlets significantly impacts the agenda-setting function of the media. As described by Vliegenthart & Walgrave (2019), the media’s agenda-setting influence is not uniform but varies according to several factors, including the nature of the issues, the type of media and the political context (p. 274). According to the media logic theory, journalists choose and convey political news based primarily on their own format needs and news values. As a result, specific content aspects of political news are emphasised (Takens J. , 2013, p. 80). Walgrave & Van Aelst (2006) suggest that media influence is stronger for issues that fit a typical media logic, which prioritises news values such as conflict, drama, and personalisation. This media logic drives the selection and framing of issues, thereby affecting their salience to the public (p. 98). Additionally, research indicates that journalists are inclined towards three content traits with significant news value: personalised coverage, contest and game coverage and negative coverage (Takens et al, 2013, 8 p. 278). In the context of this thesis, the Dutch tabloid De Telegraaf prioritises sensational and controversial aspects of policy or political figures (Sens, 2009, p. 145). Whereas NRC, a broadsheet, may provide more thorough and impartial coverage, with this being an important objective in their fundamental attitudes (Hoeven, 2012, p. 479). Hoof (2000) describes these differences in commercial and cultural qualities in newspapers. Information that meets readers' demands to the best of its ability is considered commercial quality and tabloid newspapers. Conversely, cultural quality refers to the reader's ability to learn as much as possible from the material presented. Broadsheet newspapers are experts in this area (p. 19). Broadsheets rely on an investigative approach, and their vocabulary and word choice are frequently more complex than popular ones since they aim for depth. On the other hand, tabloid content typically uses populist language to investigate and showcase current issues (Bastos, 2019, p. 2). 2.3. The Political Agenda While the media has a significant role in setting political agendas, the reverse dynamic, where political agendas shape media agendas, is equally important. Political actors often strategically engage with the media to steer public discourse in a way that aligns with their objectives (Sorensen, 2018, p. 3). Media coverage offers an opportunity for some political parties to draw attention to their top issues at the moment (Van Aelst & Walgrave, 2017, p. 6). Political parties use various tools to influence media coverage. Press releases, staged events, and social media campaigns are designed to attract media attention and frame issues in a way that supports their agenda (Arceneaux et al, 2019, p. 130). For instance, during election periods, politicians might emphasise particular policy proposals or achievements to generate favourable media coverage and shape the media agenda, which can, in turn, influence public opinion and voting behaviour (Partheymüller, 2017, p. 13). As mentioned before, the impact of digital media on the agenda-setting relationship between the media and political agendas is of interest. Gilardi et al. (2022) specified the interplay between social media and political agendas as “(…) not only do the traditional media agenda, the social media agenda of parties, and the social media agenda of politicians influence one another but, overall, no agenda leads the others more than it is led by them.” (p. 41). Highlighting the importance of social media and the dynamics in the formation of the media agendas as well as the political agendas. 9 The agenda-setting effects of the political parties are closely linked to the ownership of issues. “Issue ownership does a fair job in predicting which issues will be emphasised by which party, as well as predicting the outcomes of elections based on the dominant issues in the news.” (Meijer & Keinnijenhuis, 2006, p. 545). Political parties strategically use media platforms to assert their dominance over specific issues to align them with their ideologies and policy agendas (Camp, 2017, p. 5). This strategic issue ownership allows parties to effectively leverage both digital and traditional media to solidify their position on key political matters (Green & Jennings, 2011, p. 21). This approach underscores the power of issues ownership in political communication and its impact on the democratic process, especially in election campaigns. For example, in the Netherlands, Geers & Bos (2017) found “(…) that media exposure to issue news, linked through issue ownership, leads to vote switching.” (p. 359). Specifically, issue domination by a party is essential (p. 360). Previous research identified issues being ’owned’ by certain parties in the Netherlands. Research by Sipma & Berning (2021) on populist radical right parties, such as the PVV, shows that populist radical right parties “(…) are likely to be considered as issue owners by those who perceive immigration as one of the most important problems (…)” (p. 2). This is exemplified by a poll by I&O research with Dutch voters, with 32% of the voters thinking the party PVV has the best ideas on immigration (Kanne & van der Schelde, 2023). In the Dutch National Voter Survey 2023 (Nationaal Kiezersonderzoek), voters are asked what issues they associate with certain political parties. In a preliminary analysis of the survey by Lubbers et al. (2023), voters considered the centre-left party GroenLinks-PvdA to be the issue owner on topics regarding the climate. The liberal conservatives, VVD, are thought to have stronger issue ownership of the topic economy across groups of voters (van der Meer, 2022, p. 12). As a newcomer, the party NSC is not mentioned in previous research on issue ownership. Before the elections, the polls defined various issues connected to NSC, such as social division and poverty, but no clear issue ownership (Kanne & van der Schelde, 2023). 2.4. Professionalisation of Political Campaigns The research presented shows a dynamic interplay in the relationship between the media and political agendas. The professionalisation of political campaigns, characterised by using sophisticated media strategies, has an impact on this interplay, it can impact the formation and prioritisation of media and political agendas. According to Gibson & Römmele (2009), 10 professionalisation is a strategic approach in a diverse media environment, where political parties tailor their messages to resonate across different platforms and set the agenda (p. 266). Within this context, it is important to consider that societal changes have influenced campaign tactics. Emphasising the necessity of political campaigns to develop. Magin et al. (2017) specify four campaign types related to the relevant ages in political communication: partisan- centred campaigns (1850-1960), mass-centred (1960-1990), targeted group-centred (1990- 2008), and individual-centred (since 2008) (p. 1702). Contemporary political campaigns are characterised by a multifaceted and dynamic approach to cater for the digitally connected electorate, engaging with voters through various online platforms and utilising targeting techniques to maximise influence on the media agenda and outreach (Kreiss, 2016, p. 219). Especially in a time of increased digitalisation, practises of audience fragmentation and more personalised communication of political parties through social media are frequent (Kruikemeier et al., 2016, p. 367). In line with this, Kreis (2016) highlights the importance of integrating digital tools and platforms into campaign strategies, noting the agility and reach of social media allow for more immediate and personalised voter engagement (p. 219). One of the key aspects of professionalised political campaigns is to influence the media agenda. With professional communication teams, political actors can more effectively highlight certain issues. This professional approach often involves setting the agenda by introducing new issues or reframing existing ones in a way that benefits their political strategy (Thesen, 2013, p. 365). Research also shows that professionalised campaigns attract higher media attention (Meyer et al., 2017, p. 286), and they use various tactics to ensure their parties’ messages align with media logic, making them more newsworthy and likely to be covered (Ruthardt, 2023, p. 8). This highlights the impact of professionalisation on the political agenda. For example, by using professionalised tactics to select their key issues, political parties can strengthen their perceived expertise in certain issues (Banda, 2013, p. 447). However, unpredictability and external factors still play a significant role in campaign dynamics; Strömback & Kiousis (2014) note that events occur even with professionalised strategic-oriented campaign activities through political campaigns. “(…) election campaign communication often might be less “strategic” than it appears.” (p. 114). Professionalised campaigns in the Dutch context are no exception. The Netherlands is characterised by high political involvement and high journalistic professionalisation (Vermeer et al., 2022, p. 6). They align with global trends, employing social media platforms such as Facebook to generate organic and sponsored content to create personalised messaging and 11 foster community-building among supporters. Especially Facebook offers a relevant platform, offering many possibilities for personalisation and advertising. For example, campaigns might use Facebook to advertise short, impactful videos, live Q&A sessions with candidates, and targeted advertisements that leverage user data to reach specific demographics. Dutch campaigns actively incorporate innovative outreach methods, reflecting the shift towards professionalisation (Gibson & Römmele, 2009, p. 266). Moreover, professionalisation can also be seen in political parties using microtargeting practices on Facebook (Zuiderveen Borgesius, et al., 2018, p. 84), in which targeting practices create specific and personal campaign messages to convey a political message. Beraldo et al. (2021), in their research on Facebook advertisements in the 2021 Dutch election, acknowledge the importance of Facebook in Dutch campaign advertising, stating that many parties reach voters with targeted messages (p.1). 2.4.1. Indicators of Professionalised Campaigns To understand the role that professionalisation has on the relationship between the media and political agendas, it is important to recognise the indicators of professionalisation in campaigns. By examining these indicators in the output of political campaigns the degree of professionalisation of a political party can be researched. Gibson & Römmele (2001) presented a party-centered theory in which campaign professionalisation can be understood as a mechanism for agenda-setting. This involves using political consultants, data-driven strategies, and targeted messaging to influence public perception. Mykkänen et al. (2022) have researched the contemporary relevance of the party-centered theory. They argue that all parties and political candidates seek campaign structures and strategies to reach their goals and seek a nuanced operationalisation of the theory. “Mainstream electoral parties can be expected to rely on a more rational-oriented toolbox, while populist and extremist parties are more oriented toward emotional campaign elements.”. The difference in the tools used and definition of professionalisation should be considered” (p. 1184). Gibson & Römmele (2001) also offer specific criteria for defining a professionalised campaign. As shown in Figure 1, the Professional Campaign Index reveals a set of indicators for campaigns. 12 Figure 1 - The Professional Campaign Index (Gibson & Römmele, 2001, p. 39). The Professional Campaign Index is well-known in research but somewhat outdated. For a more contemporary implementation of the index by Gibson & Römmele (2001), Spierings & Jacobs (2019) used more recent indicators. To understand the professional use of Facebook and Twitter in the 2010 and 2012 Dutch elections by political campaigns, Spierings & Jacobs (2019) specified four types of variables of professionalisation used in the posts of parties and candidates: 1) information about their identity and policies; 2) engagement (sharing of posts, videos, etc.); 3) mobilisation (donation, get out to vote, etc.) and 4) interactions (e.g. webcare) (p. 158). The strength of these variables is supported by research conducted by Koc- Michalska et al., (2014), who underscore their importance. Their study found that engagement and mobilisation are critical indicators of a campaign's online performance, as they directly reflect its ability to interact with and activate its audience (p. 9). Visuals are also very relevant as an indicator of professionalisation. Political posts featuring images tend to be more personalised and engaging than text-only posts (Lilleker D. G., 2019, p. 40). Images and videos make the content more intimate and emotionally resonant, significantly enhancing the impact of politicians’ Facebook postings (Farkas & Bene, 2021, p. 125). This deliberate use of visual content is a key feature in understanding the efficacy of professional campaigns on social media. Additionally, the Facebook ’like’ and ‘share buttons, as well as comments, indicate political interest and engagement. For example, the frequency of ’likes’ on posts can provide insights into the reach and effectiveness of campaign messages (Heiss et al., 2018, p. 3). This integration of visual elements highlights the increasing sophistication of digital strategies, which are designed to maximise voter engagement and 13 create stronger personal connections between candidates and voters (Stromer-Galley, 2019, p. 137). Demographic targeting by political campaigns is also an indicator of professionalisation. As cited in Hirsch et al. (2024), Römmele & Gibson (2020) describe demographic targeting as a “(…) much more fine-grained and personality-based understanding of the persuadable electorate (…)” (p. 20). Showcasing deliberate use of professional campaign strategies. Dutch political parties employ online platforms to create personalised messages reflected in the use of microtargeting practises, where specific voter segments are targeted with tailored messages to maximise impact (Zuiderveen Borgesius et al., 2018, p. 84). Microtargeting practices on Facebook base the message on demographics like location, interest, age, and online behaviour (Hirsch et al., 2024, p. 555). Another strategy as an indicator of professionalisation is the money spent on political campaigns. Many parties invest a great amount of money to get advertising exposure. In this way, political parties can reach out to citizens and stay visible, thus showing their strategic commitment to increasing visibility in their campaigns (Chu, et al., 2024, p. 292). All considered, the theoretical background presented poses a well-researched relationship between media agendas and political agendas. The media and political agendas have a mutual relationship, where the media agenda can shape the political agenda and vice versa. The role professionalisation of political campaigns plays in this relationship is two-sided, enhancing political campaigns' agenda and strategic output and impacting the media agenda. Through this framework, this thesis will consider these interconnections and answer the main research question. 14 3. Methods 3.1. Research Design This study adopts a quantitative content analysis approach to examine the media agendas of a tabloid (De Telegraaf) and broadsheet (NRC), as well as the political agendas through Facebook posts and advertisements of PVV, GroenLinks-PvdA, VVD and NSC, during the final two months of the 2023 Dutch national elections. The research expands the application of the agenda-setting theory to look at the relationship between the media agendas and the political agendas and the reverse, the relationship between the political agendas and the media agendas. Although any direct causal relationship will not be concluded due to limitations in the study's design, the focus will be on the alignment and issue consistency across these mediums. This research will provide additional answers to the role of professionalisation in political campaigns on the relationship between these agendas. For this study, De Telegraaf and NRC were selected to represent the media agendas because De Telegraaf is the biggest tabloid in the Netherlands, as shown in Table 1, with an 11.2% daily reach of the Dutch population in 2021. NRC, being one of the most prominent broadsheets, has a reach of 2.6% in 2021 (Commissariaat voor de Media, 2023, p. 32). The four political parties, PVV, GroenLinks-PvdA, VVD and NSC were chosen because of their contemporary relevance in the 2023 elections, these four parties were ranked the four biggest parties after the elections. The social media platform Facebook is used for the analysis of political agendas since it is a platform where political campaign output is published, and the advertising data of the parties are available. Year 2021 NOS.nl 23.9 NU.nl 17.7 AD 13.9 De Telegraaf 11.2 RTL Nieuws 8.1 de Volkskrant 3.1 NRC 2.6 de Gelderlander 2.3 De Stentor 2.1 Indebuurt.nl nb 15 Eindhovens Dagblad 1.4 Trouw 1.7 Total Daily Brands 29.9 Total Non-Daily Brands 38.7 Table 1 - Average daily reach of the top 12 online news brands in percentages. "nb" indicates data unavailable or not measured. (Commissariaat voor de Media, 2023, p. 32). This study will rank issues and political leaders by their dominance in the coverage of tabloid De Telegraaf, broadsheet NRC and Facebook posts and advertisements of PVV, GroenLinks- PvdA, VVD and NSC through the statistical computing program R to determine if specific words were included in the articles. The software enables a precise and efficient examination of the text, facilitating the identification of relevant issues and leaders through the inclusion or absence of specific words within the dataset. Also, the issue overlaps and deviations across these mediums are traced without presupposing a fixed direction of influence and the mentions of the political leaders. Analysing the last two months of the election campaign allows for examining the media agendas and political agendas over time and will provide insight into the strategic position of parties, assessing how well they manage to ‘own’ specific issues in both the media agendas and their political agendas. The role of professionalisation in the relationship between media and political agendas is also of concern for this thesis, integrating the presence of professionalised campaign indicators. Building on the theoretical background, the six indicators as outlined by Heiss et al. (2018), Lilleker (2019), Römmele & Gibson (2020), and Jacobs & Spierings (2018) will be analysed. It is important to note that not all of the indicators may be present in the posts and advertisements of the parties. However, all available indicators will be presented. First, mobilisation practises of political parties are assessed by measuring the presence of call-to- action. Additionally, the type of media used will be analysed, highlighting the use of visual content. The engagement with the content (e.g. shares, likes and comments) is analysed, showing active voter involvement. The demographic focus, including 1) geographic location (broad and specific zip codes) and 2) age, will be specified, illustrating deliberate strategic choices in published content. Lastly, the total spending by the parties will be analysed, demonstrating the commitment of their campaigns. Through these indicators, the degree of professionalisation per political party can be established, which will offer insights into the role of professionalisation in the agenda-setting relationship between the media and political parties. 16 Initially, a sentiment analysis is considered to get a broader understanding of the tone of voice (positive, neutral and negative) in the examined articles. Natural Language Processing (NLP) allows for the interpretation and manipulation of human language by software. Specifically, the subjectivity lexicon for Dutch adjectives (Pattern.nl), as presented by De Smedt & Daelemans (2012), is considered. This lexicon includes 3000 Dutch adjectives and can index the polarity score between positive +1.0 and negative -1.0 (de Fortuyn et al., 2012, p. 11618). However, various researchers have questioned the validity of the dictionary-based approach. For instance, in a Dutch-setting comparison of methods for sentiment analysis, van Atteveldt et al. (2021) question the performance of dictionaries, with the agreement being too close to a chance agreement and correlations being low. Manual group coding and machine learning offer more reliable results (p. 128). Given these considerations, sentiment analysis was evaluated but disregarded for this analysis. It could, however, be viable for further research. 3.2. Data Collection and Processing Figure 2 - Processing steps for data analysis. 3.2.1. Sampling A purposive sampling strategy is implemented to capture the media and political agendas across selected media platforms within a specified timeframe. This method is chosen over other sampling techniques because it allows for a targeted examination of how political issues and leaders are portrayed within the media during the intense period of electoral campaigning. The sampling frame encompasses two distinct media types: newspapers (De Telegraaf and 17 NRC) and Facebook content (posts and advertisements) from the four major political parties (PVV, GroenLinks-PvdA, VVD, and NSC). Newspapers: Articles are selected if they mention any of the four political parties and leaders within the final two months leading up to the election day. This period is chosen to capture the peak of the election-related media coverage and campaigning efforts. Facebook content: Facebook posts and advertisements are selected if they are official posts made by the parties or their leaders that pertain to the election campaign, policy proposals, or direct engagement with voters within the same two-month period. 3.2.2. Data Collection Newspapers: Articles are collected using the digital archives (Nexis Uni) of De Telegraaf and NRC. Advanced search options are employed to filter articles published within the two months leading up to the election, using party names, key political leaders, and election- specific keywords as search criteria. This method ensures that the data collection respects copyright laws and ethical standards related to the use of published material. Facebook content: Data from Facebook involves scraping the official pages of the four political parties for posts and advertisements using automated methods. The Facebook Graph API is utilised to systematically extract advertisements, including metadata such as post text, date, engagement metrics (e.g. likes, shares and comments), and multimedia attachments. 3.2.3. Processing For newspaper articles and Facebook content, the process of preparing the data for analysis will be supported by R. This will involve packages designed for text processing and analysis, such as ‘tidyverse’ and ‘stringr’ for string manipulation, to automate removing irrelevant content, duplicates, stopwords and data segmentation into analysable units. R facilitates efficient and accurate processing of large datasets, ensuring that the data is prepared in a format suitable for detailed analysis and is widely used within research (Dege & Brüggeman, 2023, p. 5). In the analysis phase, R scripts are developed to categorise the textual content based on defined issues and political leaders in newspapers. These scripts prepare the data for basic descriptive analysis and advanced statistical testing. Correlation regression analyses are 18 conducted to understand the relationships between Facebook content and newspaper coverage. These analyses help identify patterns and predict the presence of certain issues based on the type of media. Additionally, text mining will be used to study textual information. Text mining refers to the process of extracting high-quality information from text. This means R can classify, summarise, cluster similar documents and extract concepts (de Fortuny et al., 2012, p. 11617). 3.3. Corpus The study focuses on the Dutch national elections in 2023, with newspaper stories released between September 22, 2023, and November 22, 2023, the election day. Newspaper articles were gathered using the Nexis Uni database. A few criteria were set for the corpus selection: to validate the applicability of this analysis, every article had to contain the terms "VVD", “NSC”, “PVV”, and/or “GroenLinks-PvdA”. Because GroenLinks-PvdA is a combined list, more combinations are possible. So, “GroenLinksPvdA”. GL/PvdA”. GL-PvdA”, “PvdA- GL”, “PvdA/GL”, “PvdA-GroenLinks”, and “PvdAGroenLinks” are added to the selection. To avoid duplications, all items are selected in Dutch and from NRC or De Telegraaf, excluding NRC.nl and DeTelegraaf.nl. After removing all duplicate and combined articles, 736 items are examined, with 429 being from De Telegraaf and 307 from NRC. Following the first examination of the articles, articles that do not contain campaign news are excluded from the examination. Table 2 specifies the total number of included newspaper articles. Table 2 – Total included newspaper articles Newspaper Frequency Percentage De Telegraaf 264 54 NRC 223 46 Total 487 100 In addition to newspapers, the corpus includes content from the official Facebook pages of the four major political parties and politicians. The selection criteria for Facebook posts mirror those of the newspaper articles, focusing on posts made between September 22, 2023, and 19 November 22, 2023. Each post included in the study is selected to directly relate to the election campaign, policy discussions, or engagement activities aimed at voters. Data collection for Facebook content involved using automated custom scraping tools to extract relevant posts, including their metadata (e.g., date, engagement metrics). A total of 376 Facebook posts and 260 advertisements are examined in the initial examination. PVV does not have a Facebook page but communicates through the party leader, Geert Wilders’ Facebook page. Thus, his posts are used for this study. NSC did not produce many Facebook posts (only 3). For this reason, the content on the Facebook page of the party leader, Pieter Omtzigt, is also included as NSC in this study. Only GroenLinks-PvdA and VVD had a Facebook ad campaign. They are included in the advertisement dataset. Table 3 lists the total included Facebook posts and advertisements per party. Table 3 – Total included Facebook posts and advertisements Political party Posts Advertisements Frequency Frequency PVV 247 GroenLinks-PvdA 40 114 VVD 64 146 NSC 25 Total 376 260 3.4. Codebook A codebook is a critical tool for ensuring systematic and unbiased analysis. It includes definitions of variables, coding rules, and examples to guide coding in accurately identifying and categorising the content (Rose et al., 2015, p. 1). For both the newspaper articles and Facebook content, the variables are derived from the literature review and research questions, focusing on issues and political leaders. The list of issues is similar to previous research on analysing issues in newspapers and other media. Specifically, a similar codebook used for a content analysis of the campaign for the Dutch Parliamentary Elections in 2006 is considered as input for the codebooks of this research (The Amsterdam School of Communications Research, 2007). The first codebook delineates criteria for identifying the presence of issues, political leaders, and other relevant variables. This codebook can be found in Appendix 1. The Facebook content is coded through a separate procedure in the second codebook. This codebook can be found in Appendix 2. 20 3.5. Reliability of the Data The reliability of the data in this study is ensured through a systematic and methodical approach to data collection, processing, and analysis. The use of a quantitative content analysis approach and a well-defined sampling strategy allows for examining the 2023 national elections in the Netherlands. Utilising digital archives and the Nexis Uni database to retrieve newspaper articles ensures that the collection process is comprehensive and replicable. The inclusion criteria for articles based on specific keywords and the election period further enhance the data's precision and relevance, contributing to internal and external validity. This precision helps to ensure that the study accurately reflects the media environment during the election period, allowing for generalisations. This study employs R for data preparation to mitigate potential biases and enhance the reliability of data processing. Using specific packages for text processing and analysis aids in systematically removing irrelevant content and duplicates, ensuring a clean dataset for analysis. This process is transparent and reproducible, as detailed in the methodology, allowing for verification and replication by other researchers, enhancing the study's credibility and contributing to its internal validity. The development and application of a codebook for both newspaper articles and social media content analysis further strengthen the reliability of this study. The codebook ensures a consistent and unbiased approach to data categorisation by defining variables, coding rules, and examples. This systematic coding process is crucial for the reliability of the analysis, as it minimises subjective interpretation and ensures that data categorisation is grounded in the study's theoretical framework. During the refinement of the codebooks, the concept of face validity, in which the degree to which the procedure appears effective in terms of its stated aims of the research, is applied to analyse the gathered data. For instance, the word ‘politie’ (police) was initially misidentified by R as part of the word ‘politiek’ (politics), demonstrating the importance of careful oversight in the coding process to maintain the integrity and validity of the dataset. 21 3.6. Analysis Next to data preparation, the data analysis will extensively utilise R's quantitative analysis capabilities. R's comprehensive library of statistical functions will be instrumental in examining the occurrence and distribution patterns of identified issues and political leaders. Specifically, a comparative framework will be implemented to analyse differences between the media agendas, the political agendas and indicators of professionalisation. To enhance the comparative aspect of the study, results will be segregated by tabloid, broadsheet, Facebook posts, Facebook advertisements, political leaders, issues and professionalisation indicators. This segmentation will aid in understanding the flow of results and their broader implications, making it easier to trace specific interactions. Additionally, tables with the number of issues included in the articles and Facebook content, heat maps of issue prevalence, and trend graphs will be used to illustrate the interconnections between the media and political agenda visually. To illustrate the professionalisation features of the political parties, various tables and bar graphs will be given to compare the parties. These visualisations will help present complex data in an accessible and interpretable manner, highlighting significant trends. For instance, heatmaps will visually depict the intensity and distribution of issues across different media and political parties, offering a representation of how different issues are prioritised. Similarly, a week-by-week trend analysis conducted on the media and political agendas will reveal how issues gain or lose prominence over time. This analytical strategy, emphasising comparative metrics, will provide a framework for interpretation. 22 4. Results This chapter presents the findings from the analysis of the relationship between the media agenda in Dutch newspapers De Telegraaf and NRC and the political agendas of four major political parties (PVV, GroenLinks-PvdA, VVD and NSC) during the final two months of the 2023 Dutch national elections. Additionally, it will give insights into the role professionalisation of political campaigns has on the interaction between these agendas. The results are structured to respond directly to the research questions outlined in the introduction. 4.1. Issues and Political Leaders in the Media Agendas 1. Which issues and political leaders are most prominently featured in the media agendas in the last two months of the 2023 Dutch national elections? Table 4 shows the top 10 most mentioned issues in De Telegraaf and NRC, the total number of articles that mention the issue, and the percentage of relative occurrence compared to the total number of articles. Table 4 – Top 10 Issues in NRC and De Telegraaf Rank Issue NRC De Telegraaf NRC De Telegraaf Mentions Percentage Mentions Percentage 1. 1. Nature and Environment 103 46.2 81 30.7 2. Education and Science 75 33.6 3. 8. General Economic 66 29.6 31 11.7 Policy 4. 4. Road Traffic 60 26.9 34 12.9 5. 10. Housing 55 24.7 24 9.1 6. 2. Immigration 54 24.2 52 19.7 7. 6. Public Transport 46 20.6 32 12.1 8. 5. Sport 44 19.7 34 12.9 9. European Union 38 17 10. Wages and Collective 38 17 Labour Agreements 3. Taxes 36 13.6 7. Israel-Palestine Conflict 31 11.7 23 9. Safety 24 9.1 The findings on the top 10 issues in the newspapers indicated a variation in coverage by both newspapers. De Telegraaf showcased many articles on nature and environment (81 articles, 30.7%) and immigration (52 articles, 19.7%). In comparison, NRC highlighted nature and environment (103 articles, 46.2%) and education and science (75 articles, 33.6%). Next to the similarities on the issue nature and environment, the results indicate a difference in the media agendas between De Telegraaf and NRC. NRC's coverage is characterised by a stronger emphasis on long-term societal issues, with nature and environment and education and science receiving significantly more attention. In contrast, De Telegraaf provides a more varied spread, suggesting an additional editorial focus on current, tangible issues like taxes, road traffic, and public transport. Secondly, Table 5 presents the total number of mentions of the four party leaders in De Telegraaf and NRC and the relative occurrence compared to the total number of articles in both newspapers. Table 5 – Mentions of Political Leaders in De Telegraaf and NRC Rank Political leader NRC De Telegraaf NRC De Telegraaf Mentions Percentage Mentions Percentage 1. 1. Pieter Omtzigt 119 53.6 108 40.9 2. 2. Frans Timmermans 80 35.9 93 35.2 3. 3. Geert Wilders 63 26.9 48 18.2 4. 4. Dilan Yeşilgöz-Zegerius 54 24.2 44 16.7 Regarding the portrayal of political leaders, Pieter Omtzigt and Frans Timmermans were considerably more mentioned in both De Telegraaf and NRC, with Omtzigt receiving the highest percentage of mentions (40.9% in De Telegraaf and 53.6% in NRC). Omtzigt's higher percentage of mentions in NRC could indicate that his political platform or activities resonate more with the issues that NRC readers are interested in or that he was involved in more newsworthy events during the period studied. 24 When looking at sub-question 1a: How do the media agendas differ between broadsheet (NRC) and tabloid (De Telegraaf)? A Spearman’s rank-order correlation (as presented in Table 6) was conducted to assess the association between the rankings of issues in the coverage by NRC and De Telegraaf. Results indicated a weak positive correlation between the rankings of issues in the two newspapers, ρ = 0.6. This result suggests a moderate positive correlation between the rankings of issues within the article coverage of the two newspapers during the final two months of the 2023 campaign. Despite this, the significance level is just above a standard threshold for statistical significance (0.05), suggesting that the correlation observed is not statistically significant. Table 6 – Spearman’s rank-order Correlation Issues in NRC and De Telegraaf Issues in NRC Issues in De Telegraaf Spearman’s rho Issues in NRC Correlation Coefficient 1.000 .6 Sig. (2-tailed) .07 N 10 10 Issues in De Correlation Coefficient .6 1.000 Telegraaf Sig. (2-tailed) .07 N 10 10 * Correlation is significant at the 0.05 level (2-tailed) To better understand the media agendas between the tabloid and broadsheet, two graphs are provided, which depict a week-by-week trend analysis of the ten most frequently discussed issues in NRC and De Telegraaf. These trend lines show the issue coverage over time, visually representing how specific issues gain longitudinal or wane attention as the election approaches. This descriptive analysis is critical for understanding patterns in the data, revealing how issues develop over time and in comparison between the newspapers. In Table 7, the weekly distribution used for analysis is specified. Table 7 – Weeks distribution Weeks Dates Week 1 22-09-2023 until 28-09-2023 Week 2 29-09-2023 until 05-10-2023 25 Week 3 06-10-2023 until 12-10-2023 Week 4 13-10-2023 until 19-10-2023 Week 5 20-10-2023 until 26-10-2023 Week 6 27-10-2023 until 02-11-2023 Week 7 03-11-2023 until 09-11-2023 Week 8 10-11-2023 until 16-11-2023 Week 9 17-11-2023 until 22-11-2023 In the trend analysis of NRC (Graph 1), the coverage of the issue nature and environment showed a marked and consistent increase, leading up to a significant peak in the last two weeks, suggesting a growing emphasis on environmental issues or events related to nature. education and science also demonstrated an upward trend, although with more weekly fluctuations, indicative of a sustained interest in this issue. Other issues, such as road traffic, housing, public transport, and the European Union, exhibited relatively stable coverage with minor fluctuations throughout the period. Immigration saw a stark increase around week 8, indicating a possible event-driven spike in coverage. In contrast, sport, and wages and collective labour agreements received the least attention, remaining relatively low throughout the nine weeks. Graph 1 – Trend analysis of issues in NRC 35 30 25 20 15 10 5 0 Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 Nature and Environment Education and Science General Economic Policy Road Traffic Housing Immigration Public Transport Sport European Union Wages and Collective Labour Agreements 26 Mentions in articles The trend analysis of De Telegraaf (Graph 2) shows a different pattern of interest. The issue nature and environment also increased overall, although with some weekly variability. An interesting trend is the peaks around immigration coverage weeks 4 and 8. Also, the issue of public transport saw a sharp increase towards week 8, possibly reflecting a significant event or policy change. Other issues, such as immigration, general economic policy, taxes, and safety, did not follow a clear trend, fluctuating throughout the weeks without a discernible pattern. This could suggest sporadic reporting influenced by current events rather than a sustained focus. Interestingly, De Telegraaf featured coverage of the Israel-Palestine conflict and sport, issues that either did not appear or were less emphasised in NRC. The Israel-Palestine conflict increased significantly in week four and declined sharply in week 6, indicating heightened interest in this period (just after the conflict started on October 7th 2023). Graph 2 – Trend analysis of issues in De Telegraaf 18 16 14 12 10 8 6 4 2 0 Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 Nature and Environment Immigration Taxes Road Traffic Sport Public Transport Israel-Palestine Conflict General Economic Policy Safety Housing Comparing the two graphs, it is apparent that while both newspapers extensively covered nature and environment, their reporting patterns on other issues diverged. This may reflect different editorial priorities or reader interests. For instance, the specific increases in 27 Mentions in articles immigration in De Telegraaf may indicate a targeted interest or event that was particularly relevant to its audience. Next, an analysis of the relationship between the issues and political leaders in De Telegraaf and NRC is given (Graphs 3 and 4). The values in the following graphs are relative to the total mentions of the issues and political leaders in the newspapers for comparison purposes. Graph 3 – Heat Map Issues vs. Political Leaders in NRC Nature and Environment 36 51 26 57 Education and Science 22 22 16 40 General Economic Policy 18 21 12 32 Road Traffic 20 23 15 30 Housing 14 19 12 28 Immigration 27 25 20 36 Public Transport 10 14 7 22 Sport 14 19 13 26 European Union 11 15 6 17 Wages and Collective Labour Agreements 12 17 8 27 Geert Frans Dilan Yeşilgöz- Pieter Wilders Timmermans Zegerius Omtzigt Graph 4 – Heat Map Issues vs. Political Leaders in De Telegraaf Nature and Environment 20 36 21 37 Immigration 18 19 12 24 Taxes 4 14 6 17 Road Traffic 7 14 7 16 Sport 5 15 6 18 Public Transport 9 11 5 13 Israel-Palestine Conflict 4 16 3 9 General Economic Policy 6 11 8 16 Safety 3 4 5 9 Housing 3 14 4 9 Geert Frans Dilan Yeşilgöz- Pieter Wilders Timmermans Zegerius Omtzigt 28 Issues Issues The heat maps show that the four political leaders have different associations with the issues presented in De Telegraaf and NRC. For instance, Geert Wilders of the PVV is strongly linked to the issue of nature and environment and immigration in both newspapers, aligning the parties' issue ownership on immigration. Frans Timmermans, representing GroenLinks-PvdA, shows a balanced association across several issues, but with a notable emphasis on the nature and environment in NRC and De Telegraaf, reflecting his involvement with climate issues. Dilan Yesilgöz-Zegerius of the VVD demonstrates no domination on one specific issue. Lastly, Pieter Omtzigt, associated with the NSC, has a link to most issues in both newspapers, although having more relation to nature and environment, education and science and immigration, indicating the broad range of issues in the newspapers. This array of associations between political leaders and issues underscores the diversity of their political engagements and the media's role in highlighting these connections to the public. Conclusion In exploring media agendas, Spearman’s rank-order correlation suggested a moderate positive correlation with the rankings of issues between the two newspapers. The significance level is just above a standard threshold for statistical significance (0.05), suggesting that the correlation observed is not statistically significant. However, what the findings do show from the media agendas of De Telegraaf and NRC , De Telegraaf, balancing between covering pressing societal issues, such as nature and environment and immigration, and issues more directly relevant to everyday life, such as taxes and public transport. This editorial focus could play into the concept of issue ownership, where De Telegraaf highlights issues commonly associated with particular political parties. Interesting trends in De Telegraaf are the specific peaks in immigration coverage, possibly reflecting specific events or policy changes as well as the consistent mentions of nature and environment, showcasing a growing emphasis on the issue. The issue of public transport also sharply increased coming closer to election day. De Telegraaf also featured coverage of the Israel-Palestine conflict, however, this is probably due to heightened interest during this period. Conversely, NRC emphasises long-term societal issues, with education and science and nature and environment leading its coverage. The focus on these subjects aligns with NRC's objectives in providing more thorough content that informs. In the trend analysis, the coverage of nature and environment showed an increase, reaching a significant peak in the 29 last two weeks, similar to De Telegraaf, suggesting a growing emphasis on the issue. Immigration and education and science coverage also saw an increase towards election day, indicating possible event-driven spikes. Analysing political leaders' media presence reveals that Pieter Omtzigt and Frans Timmermans have the most significant share of attention in both newspapers. Omtzigt, in particular, is receiving greater emphasis in NRC. The substantial coverage in Omtzigt aligns with a broader, less-defined issue spectrum, perhaps enabling him to step outside the traditional boundaries of issue ownership. Timmermans' frequent mentions, especially in discussions on environmental matters, align with issue ownership as presented in the theoretical background, where his party is seen as the issue owner of environmental and climate-related discourse. The heat maps clarify these dynamics. Other leaders, such as Geert Wilders and Dilan Yeşilgöz-Zegerius, also show varied levels of issue association but with less frequency and intensity than Omtzigt and Timmermans. Wilders, for instance, is notably linked with immigration in both newspapers, which aligns with his and his party’s political focus. However, this coverage is less pervasive than that of Omtzigt and Timmermans. The differential emphasis on various leaders and issues reveals the media agendas and underscores the nuanced role of the media in either reinforcing or challenging the traditional boundaries of issue ownership among Dutch political parties. 30 4.2. Issues and Political Leaders in the Political Agendas 2. Which issues and political leaders are most prominently featured in the political agendas in the last two months of the 2023 Dutch national elections? Now that the media agendas are presented, the political agendas of PVV, GroenLinks-PvdA, VVD and NSC are given. The content of the Facebook campaign posts and advertisements is examined more closely and presented in Table 8 with the percentages relative to the total posts or advertisements published by the party in question. When considering the issues mentioned in posts by the four parties, it becomes evident that some had specific focal issues. However, it is essential to mention that for VVD (16% mentioned issues in posts) and NSC (32% mentioned issues in posts), there was no considerable amount or specific emphasis on the issues mentioned. The posts of NSC were more videos and images without any text attached. Thus, for most posts, no text-based issues could be extracted. The same can be said about VVD's advertisements, which only mention issues in 8%. GroenLinks-PvdA posts (53% mentions of issues) and their advertisements (31% mentions of issues) concentrated on nature and environment, aligning with their ‘owned’ issue, showing consistency in ownership of the issue. PVV (36% mentions of issues) spread their messaging across posts with a broader range of issues, with a noticeable emphasis on the Israel-Palestine conflict and immigration. These issues resonate with their party platform on national identity and social policy. Table 8 – Issues mentioned in posts and advertisements of PVV, GroenLinks-PvdA, VVD and NSC in percentages Issue PVV GroenLinks-PvdA VVD NSC Posts Posts Advertisements Posts Advertisements Posts Nature and 2 35 31 5 1 Environment Immigration 12 General Economic 4 3 1 4 Policy Israel-Palestine 15 Conflict Taxes 2 3 2 Housing 5 6 European Union 2 31 Youth Policy 3 2 Emancipation 5 Integration 3 1 Public Transport 5 4 Sport 3 5 Participation 12 Social Division 5 4 4 Ethical Issues 4 Wages 3 Education and 3 Science Total 36 53 58 16 8 28 Table 9 compares how often political leaders are mentioned in the content of their respective parties in relative percentages to the total posts and advertisements of the party. The VVD stands out with a highly concentrated strategy, featuring Dilan Yesilgöz-Zegerius in 86% of their posts and 73% of their ads, indicating a focused promotional effort on these figures. GroenLinks-PvdA prefers Frans Timmermans in their posts at 30% but rarely highlights any political leaders in their ads, suggesting a broader party-centric approach in advertising. Notably, PVV and NSC mention their political leaders, Geert Wilders and Pieter Omtzigt, relatively infrequently in posts at 12% and 8%, respectively, and even less in advertising. This low percentage could indicate a strategic choice to emphasise party identity or collective leadership over individual personalities. Table 9 – Political leaders mentioned in posts and advertisements of PVV, GroenLinks- PvdA, VVD and NSC in percentages Politic PVV GroenLinks-PvdA VVD NSC Posts Posts Advertisements Posts Advertisements Posts Pieter Omtzigt 1 8 24 Frans Timmermans 3 30 8 2 8 Geert Wilders 12 3 1 32 Dilan Yeşilgöz- 4 3 86 73 Zegerius Total 20 44 9 88 73 32 To understand the political agendas better over time, three trend graphs of the posts and advertisements of the four parties are shown. They depict a week-by-week trend analysis of the three most frequently discussed issues in the newspapers. The three most frequent issues are discussed because of the relatively low occurrences of issues in the political agendas overall. The same weekly distribution as presented for the newspapers in Table 7 is used. In Graph 5 the analysis shows distinct peaks in the mentions of the nature and environment issue. The ads from GroenLinks-PvdA exhibit significant peaks around weeks 5 and 7, likely reflecting a targeted campaign or an event response. This contrasts with the GroenLinks-PvdA posts, which also show a small but consistent amount of mentions, suggesting a different focus in their advertising strategy. PVV and VVD posts maintain lower and unstable mentions, indicating a less pronounced focus on the issue. NSC's mentions remain flat throughout, indicating no engagement with this issue Graph 5 – Trend analysis issue nature and environment in posts and advertisements of PVV, GroenLinks-PvdA, VVD and NSC 15 10 5 0 Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 PVV posts GroenLinks-PvdA posts GroenLinks-PvdA ads VVD posts VVD ads NSC Graph 6 illustrates an upward trend in the mentions of immigration in PVV posts, which progressively increase from week 1 to week 9. This suggests a growing emphasis or campaign focus on immigration issues. All other parties have no mentions of immigration in their posts and advertisements. 33 Mentions in posts/ads Graph 6 – Trend analysis issue immigration in posts and advertisements of PVV, GroenLinks-PvdA, VVD and NSC 8 6 4 2 0 Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 PVV posts GroenLinks-PvdA posts GroenLinks-PvdA ads VVD posts VVD ads NSC Graph 7 mentions general economic policy across different platforms and shows notable variability. VVD posts and ads show one spike in week 6, which may reflect responses to specific economic events or policy announcements. NSC shows a similar result with a mention of the issue in week 7. The posts from PVV show fluctuations but without a clear trend, suggesting sporadic attention to economic issues rather than a sustained campaign focus. GroenLinks-PvdA's line shows a sharp increase in the last weeks, potentially indicating a strategic shift to emphasise economic policies towards the campaign's end. Graph 7 – Trend analysis issue general economic policy in posts and advertisements of PVV, GroenLinks-PvdA, VVD and NSC 4 3 2 1 0 Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 PVV posts GroenLinks-PvdA posts GroenLinks-PvdA ads VVD posts VVD ads NSC 34 Mentions in posts/ads Mentions in posts/ads Conclusion The findings reveal the parties’ agenda during the last two months of the 2023 Dutch national elections. GroenLinks-PvdA foregrounds the nature and environment issue in their posts and advertisements, demonstrating consistent issue ownership and likely aiming to reinforce their environmental credentials with voters. Their campaign saw significant peaks around specific weeks, indicating targeted efforts. In contrast, VVD displays an absence of any focused issue discussion in their posts and advertisements, hinting at a strategic decision for the personalisation of their political leader. In the trend analysis, VVD showed spikes on the issue of general economic policy in their posts and advertisements, with much variability, likely in response to specific economic events or policy announcements. NSC also showed an absence of issues in their posts and lacked a clear trend in the trend analysis. However, within their Facebook content, NSC used many forms of visual engagement (such as videos, which are not analysed in this thesis) over the textual elaboration of issues. PVV presents a more diversified issue spread, with posts touching on various issues, including a notable emphasis on immigration and the Israel-Palestine conflict and their mentions of immigration show a progressive increase over the campaign period in the trend analysis, reflecting their established ‘owned’ issue. Regarding referencing political leaders, VVD's approach is markedly individual-focused, with a very high percentage of content spotlighting Dilan Yesilgöz-Zegerius, which could indicate a personality-driven campaign strategy. GroenLinks-PvdA, while often mentioning Frans Timmermans in posts, does not focus on individual political leaders in advertisements, perhaps to emphasise a collective party image. The infrequent mention of Geert Wilders by the PVV may show their concentration on specific issues in their content. The low mentions of Pieter Omtzigt by the NSC could also be a strategic attempt to dilute the personalisation of politics in favour of a broader party-centric narrative. Notably, the overall low incidence of issue mentions, particularly by VVD and NSC, may reflect a strategic move to keep the campaign's digital content action-oriented, possibly focusing on the visual and emotional appeal rather than detailed policy discussions. 35 4.3. Professionalisation of Political Campaigns 3. What are the professionalisation practises of the political parties, and to what extent do they differ in their degree of professionalisation in the last two months of the 2023 Dutch national elections? Now that media and political agendas are examined, to what degree the political parties have professionalised campaigns and in what ways they differ will be explored. First, based on the six professionalisation indicators as mentioned in the theoretical background, the indicators available in the posts and advertisements of political parties on Facebook will be shown. Then the differences between the parties will be presented concerning their degree of professionalisation. It is important to note that the Facebook posts are retrieved from all parties, but only GroenLinks-PvdA and VVD have paid Facebook advertisements. The results are presented on posts and advertisements separately. One of the indicators presented is the call-to-action in Table 10, in which voters are encouraged to go vote and share the party's message as a way of mobilisation. Because of a difference in the number of Facebook posts and advertisements used, a percentage of the relative occurrence is given compared to the total number of posts of the party in question. Table 10 – Call-to-action of PVV, GroenLinks-PvdA, VVD and NSC in posts and advertisements Political party Posts Advertisements Total Percentage Total Percentage PVV 106 42.9 GroenLinks-PvdA 16 40 65 57 VVD 6 9.4 11 7.5 NSC 5 20 Call-to-action is a significant aspect of professionalised content. It directly encourages engagement and participation from the audience. PVV leveraged this feature extensively in the posts, with 42.9% of their posts including a call-to-action, followed by GroenLinks-PvdA at 40%. VVD and NSC used this strategy considerably less, at 9.4% and 20%, respectively. A similar outcome is seen with advertisements, as GroenLinks-PvdA has a call-to-action in 57% of its advertisements and VVD 7.5%. This indicates a disparity in the parties' engagement 36 strategies, with PVV and GroenLinks-PvdA demonstrating a more proactive approach to mobilising voters through Facebook. Media use, such as images and videos, is another indicator of professionalisation, potentially increasing the appeal and shareability of posts. The data (Graph 3) shows that VVD and NSC are inclined towards video-based posts, with 56% and 58% having no text-based posts. The same can be said of GroenLinks-PvdA, but the inclination is more towards images with 55% of the posts. The PVV used all visual content options, with the most prominent use of images making up 41% of their posts. Graph 3 – Type of media used in Facebook posts of PVV, GroenLinks-PvdA, VVD and NSC in percentages P V V 19% 26% 14% 41% Gr o e n Li n k s - P v d A 45% 55% V V D 56% 44% N S C 58% 42% Attachment Video Text Image Moving on to the parties who also had advertisements (Graph 4), a distinct trend favouring video content is observed, especially with VVD, which employs videos in 78% of their advertisements, and GroenLinks-PvdA employs videos in 47% of their advertisements, indicating a strong strategic focus on this media type for engagement. At the same time, GroenLinks-PvdA shows more diversification in their media usage, with images (15%) and multimedia (38%), compared to VVD’s images (19%) and multimedia (3%) comprising their advertisement content. This suggests a strategy aimed at capturing attention through various content forms. 37 Graph 4 – Type of media used in Facebook advertisements of GroenLinks-PvdA and VVD in percentages G r o e n Li n k s - P v d A 38% 47% 15% V V D 3% 78% 19% Multimedia Video Text Image The engagement data, as reflected in all four parties' comments, shares, and likes of the posts, is another indicator and provides their online presence. This data, presented in Table 11, offers a view of the parties’ average engagement per post, showcasing an audience's reach on a subsequent basis. Table 11 – Post engagement of PVV, GroenLinks-PvdA, VVD and NSC Political party Total comments, Average engagement shares and likes per post PVV 963781 3902 GroenLinks-PvdA 16350 409 VVD 36150 565 NSC 8753 350 The PVV outperforms with a total of 963,781 interactions and an average of 3,902 per post, pointing to a visible online strategy and a highly engaged supporter base. GroenLinks-PvdA, although with a lower total engagement at 16,350, maintains a relatively high average engagement of 409 per post. Also, the VVD’s engagement is 36,150 total interactions and an average of 565 per post. Suggesting active participation from followers of both GroenLinks- PvdA and VVD. With a significantly lower engagement of 8,753 total interactions and an average of 350 per post, NSC shows a relatively low engagement. NSC is a newcomer in the political arena, and people might not have found the party's Facebook pages yet. 38 Moreover, the advertisements by GroenLinks-PvdA and VVD show more indicators of professionalisation in the retrieved data. First of all, is the advertisement segmentation. Table 12 shows the two parties’ general (broad demographics) versus segmented (specific demographics) targeting practices. Table 12 – Segmentation of advertisements of GroenLinks-PvdA and VVD in percentages Political party Specific Broad demographics demographics GroenLinks-PvdA 54 46 VVD 86 14 The data reveals that the advertising strategies of GroenLinks-PvdA and VVD differ substantially in segmentation. GroenLinks-PvdA's approach is more balanced, with a nearly even split between targeting broad demographics (46%) and specific demographics (54%), suggesting a dual strategy aiming to reach a broad audience while concentrating on particular groups. In contrast, the VVD heavily favours specific demographic targeting, with 86% of its advertisements tailored to specific segments. It indicates a highly focused approach that could reflect a strategic decision to appeal to niche audiences with precise messaging. This difference in segmentation strategies may reflect the parties' campaigning tactics and underlying philosophies on engaging with the electorate. Targeting practises based on age is also an indicator of professionalisation practises. In Table 13, the targeting practices based on age demographics are shown. Table 13 – Age demographic targeting of advertisements of GroenLinks-PvdA and VVD in percentages Political party Targeting based on Broad targeting age GroenLinks-PvdA 5 95 VVD 0 100 The data in Table 13 shows that GroenLinks-PvdA displays a modest approach to age targeting, directing only 5% of their advertisements towards specific age groups while 39 maintaining a broad appeal with the remaining 95%. This could imply an attempt to maintain a universal message that resonates across generations. Conversely, VVD does not engage in age-specific targeting, with their advertisements focused on broad targeting. This approach may reflect a strategy to persue a wide net, aiming to connect with individuals of all ages without age-related segmentation in their messaging. Lastly, Table 14 shows the parties’ professionalisation through their spending on social media campaigns. The total spending and average per advertisement indicate the parties’ financial investment in their online presence. Table 14 – Advertisement spending of GroenLinks-PvdA and VVD Political party Total spending Average spending per (in euros) advertisement (in euros) GroenLinks-PvdA 131.749 1156 VVD 42.275 290 The data in Table 14 portrays a significant difference in the advertisement spending habits of GroenLinks-PvdA compared to the VVD. GroenLinks-PvdA’s total expenditure on advertisements is a substantial €131,749, with an average spending of €1,156 per advertisement, indicative of a strategy that invests heavily in the quality and perhaps the targeting precision of their ads. On the other hand, the VVD has a more modest total spending of €42,275 with an average of €290 per advertisement, suggesting a more economical approach or a focus on fewer, broader campaigns. This stark contrast in spending habits between the two parties may reflect differing priorities in their campaign strategies, with GroenLinks-PvdA potentially valuing a more comprehensive reach or more sophisticated ad campaigns. In contrast, the VVD might be focusing on other forms of outreach. 40 Conclusion The results of the indicators of professionalisation in the Facebook campaign content reveal the degree of professionalisation employed by PVV, GroenLinks-PvdA, VVD, and NSC. The utilisation of calls-to-action, a key indicator of professionalised political communication, varied significantly, with PVV and GroenLinks-PvdA leveraging this approach more effectively in posts and advertisements. This variance underscores the parties' strategic efforts to mobilise their electorate through direct engagement. Moreover, the deployment of different media forms in the campaign content, particularly the preference for video content in VVD’s and NSC’s Facebook posts, exhibits an understanding of the platform's dynamics to enhance message reach and voter appeal. The advertisement expenditure further illustrates the commitment to professionalised campaigning, especially by GroenLinks-PvdA, whose spending significantly exceeded that of the VVD, indicating a deliberate investment in targeted communication. Engagement metrics, such as comments, shares, and likes, illuminate the degree of interaction between the parties and the electorate. The PVV’s engagement numbers suggest a resonant campaign, while the NSC’s more modest figures highlight the challenges newer parties may face in establishing a digital foothold. The segmentation of advertisements, particularly the specific demographic targeting by the VVD, alongside GroenLinks-PvdA's more balanced approach, reflects strategic choices in voter outreach. These choices may well influence the parties' ability to connect with specific segments of the electorate. It is interesting, however, that by far, most advertisements are targeted broadly regarding age. This questions how the campaigns considered Facebook as a tool. Young people, for example, are not explicitly targeted. This could be because young people are more active on other social media platforms that are not included in this research (e.g., Instagram). When considering the degree of professionalisation between the four parties, GroenLinks- PvdA emerges as the only party that has applied all six indicators effectively. They demonstrated effective use of calls to action, invested significantly in advertisement spending, and maintained a balanced approach to demographic targeting, including specific and broad segments The VVD follows by applying all six indicators, however, showing little call-to- action practises, not utilising the mobilisation opportunities. At the same time, together with GroenLinks-PvdA, they are the only party having advertisements on Facebook. Next, PVV’s use of call-to-action and high engagement levels indicate a highly professional approach to digital campaigning. Their lack of advertisement and demographic differentiation might 41 suggest a less refined approach. The party NSC has a low degree of professionalisation, having no advertisements, showing low mobilisation practises through call-to-action, and having the lowest engagement. By knowing the degree of professionalisation of these political campaigns, the next step is to see if parties with higher or lower professionalisation differ in the correlation outcomes between the media and political agendas. 42 4.4. Relationship between Political and Media Agendas 4. To what extent do the political agendas and the media agendas share the same agenda in the last two months of the 2023 Dutch national elections, and what role does professionalisation have in the interaction between these agendas? The last step in understanding the media and political agendas and the role of professionalisation within the relationship of these agendas is to analyse if the agendas correlate. Pearson’s correlation coefficients were computed for the top three issues for both newspapers to understand the relationship between media agendas and political agendas. This is done by correlating the week-by-week mentions in articles, posts, and advertisements. Tables 15, 16, 17 and 18 show the coefficients for the issues that are present in the Facebook content of the respective party and newspaper. First, correlation coefficients for NRC are presented. For the Facebook posts, Table 15 reveals a significant positive correlation between the frequency of the issues nature and environment and immigration in PVV's Facebook posts and the coverage in NRC, with r = .869, p = .002, and r = .717, p = .003, respectively. These results indicate that the political agenda of PVV and the media agenda of NRC align. This could suggest a parallel focus or mutual reinforcement of these issues. For GroenLinks-PvdA, no significant correlation was observed in Facebook posts concerning nature and environment (r = .428, p = .251) and general economic policy (r = .179, p = .646), nor was there a significant correlation for general economic policy (r = .607, p = .083) in NSC's posts. VVD’s Facebook posts showed a negative, yet not significant, correlation of nature and environment (r = -.112, p = .754) and no significant correlation of general economic policy, which suggests no substantive relationship between the political agendas of the parties' posts on these issues and the media agenda of NRC. 43 Table 15 – Pearson’s correlation Facebook posts PVV, GroenLinks-PvdA, VVD and NSC with NRC Issue PVV Issue GroenLinks-PvdA r p SD M r p SD M Nature and .869 .002* .726 .556 Nature and .428 .251 1.509 1.556 Environment Environment Immigration .717 .003* 1.787 3.222 Immigration General .557 .119 .726 .556 General .179 .646 .333 .111 Economic Economic Policy Policy Issue VVD Issue NSC r p SD M r p SD M Nature and -.112 .754 .5 .333 Nature and Environment Environment Immigration Immigration General .071 .855 .667 .222 General .607 .083 .333 .111 Economic Economic Policy Policy Note. r = Pearson correlation coefficient; SD = Standard Deviation; M = Mean; p = p-value. Correlation coefficients range from -1 (perfect negative correlation) to +1 (perfect positive correlation), with 0 indicating no correlation. Values reported are to three decimal places. Empty spaces mean no data is available. P-values less than .05 indicate statistically significant correlations. * Correlation is significant at the 0.05 level (2-tailed) The correlation analysis of Facebook advertisements for GroenLinks-PvdA and VVD with NRC coverage, as shown in Table 16, yielded no significant correlations across the examined issues. For GroenLinks-PvdA, r = .019, p = .962, for the issue nature and environment and r = .038, p = .319 for the issue general economic policy indicate no significant relationship between their advertisement content on these issues and NRC's narrative. Similarly, VVD's Facebook advertisements showed no significant correlation with NRC for the issues of nature and environment (r = -.060, p = .877) and General Economic Policy (r = .071, p = .855). 44 Table 16 – Pearson’s correlation Facebook advertisements GroenLinks-PvdA and VVD with NRC Issue GroenLinks-PvdA Issue VVD r p SD M r p SD M Nature and .019 .962 4.833 3.889 Nature and -.060 .877 .441 .222 Environment Environment Immigration Immigration General .038 .319 1.014 .444 General .071 .855 .333 .111 Economic Economic Policy Policy Note. r = Pearson correlation coefficient; SD = Standard Deviation; M = Mean; p = p-value. Correlation coefficients range from -1 (perfect negative correlation) to +1 (perfect positive correlation), with 0 indicating no correlation. Values reported are to three decimal places. Empty spaces mean no data is available. P-values less than .05 indicate statistically significant correlations. * Correlation is significant at the 0.05 level (2-tailed) These findings suggest that while some congruence exists between PVV’s Facebook posts and NRC’s issue emphasis, particularly for nature and environment and immigration, this pattern does not hold consistently across other parties and issues. The lack of significant correlation for Facebook advertisements indicates that the impact of paid content on newspaper coverage may be limited or overshadowed by other factors not captured in this analysis. Similarly, the correlation coefficients for the Facebook posts, advertisements and De Telegraaf are presented. The results for Facebook posts and De Telegraaf, as presented in Table 17, show varying degrees of association. A statistically significant positive correlation was found between the issue of nature and environment in the PVV’s Facebook posts and its coverage in De Telegraaf, r = .777, p = .014, implying alignment between PVV’s political agenda and De Telegraaf’s media agenda on this issue. The issue of immigration, while discussed in PVV's posts, did not show a significant correlation, r = .437, p = .239, which suggests that De Telegraaf's coverage of immigration may not be as closely linked to PVV’s postings on the platform. GroenLinks-PvdA’s, VVD's and NSC’s Facebook posts showed no significant correlation with De Telegraaf's coverage for any of the issues analysed, including the issues of nature and environment and general economic policy. This lack of significant correlation indicates that 45 the political agendas of the parties do not have a discernible relationship with the media agenda in De Telegraaf on these issues. Table 17 – Pearson’s correlation Facebook posts PVV, GroenLinks-PvdA, VVD and NSC with De Telegraaf Issue PVV Issue GroenLinks-PvdA r p SD M r p SD M Nature and .777 .014* .726 .556 Nature and .577 .104 1.509 1.556 Environment Environment Immigration .437 .239 1.787 3.222 Immigration General -.197 .612 .726 .556 General .493 .177 .333 .111 Economic Economic Policy Policy Issue VVD Issue NSC r p SD M r p SD M Nature and .141 .717 .5 .333 Nature and Environment Environment Immigration Immigration General .493 .177 .667 .222 General .107 .784 .333 .111 Economic Economic Policy Policy Note. r = Pearson correlation coefficient; SD = Standard Deviation; M = Mean; p = p-value. Correlation coefficients range from -1 (perfect negative correlation) to +1 (perfect positive correlation), with 0 indicating no correlation. Values reported are to three decimal places. Empty spaces mean no data is available. P-values less than .05 indicate statistically significant correlations. * Correlation is significant at the 0.05 level (2-tailed) When examining the correlations for Facebook advertisements, the results presented in Table 18 indicate that GroenLinks-PvdA's advertisement content on general economic policy showed a moderate yet non-significant correlation with De Telegraaf, r = .522, p = .150. Also, no significant correlation was found for nature and environment, r = .204, p = .598. For VVD as well, there was no significant correlation between their advertisements' content for the issues of nature and environment and general economic policy and De Telegraaf’s coverage. 46 Table 18 – Pearson’s correlation between Facebook advertisements GroenLinks-PvdA and VVD with De Telegraaf Issue Groenlinks-PvdA Issue VVD r p SD M r p SD M Nature and .204 .598 4.833 3.889 Nature and .107 .785 .441 .222 Environment Environment Immigration Immigration General .522 .150 1.014 .444 General .493 .177 .333 .111 Economic Economic Policy Policy Note. r = Pearson correlation coefficient; SD = Standard Deviation; M = Mean; p = p-value. Correlation coefficients range from -1 (perfect negative correlation) to +1 (perfect positive correlation), with 0 indicating no correlation. Values reported are to three decimal places. Empty spaces mean no data is available. P-values less than .05 indicate statistically significant correlations. * Correlation is significant at the 0.05 level (2-tailed) These results suggest that for only the PVV, with the issue nature and environment, there is a significant alignment between their political agenda and De Telegraaf’ media agenda on this issue. The data does not substantiate such a relationship for GroenLinks-PvdA, VVD, and the NSC. This may imply that while some political agendas are reflected in media agendas, others maintain a distinct separation, pointing towards a more complex interaction. Conclusion The Pearson’s correlation coefficients reveal interesting insights between media agendas and political agendas. For the correlations between the political agendas and the media agenda of NRC, there are significant positive correlations with the issues nature and environment and immigration in the Facebook posts of the PVV, showing alignment between the political agenda of the PVV and the media agenda of NRC. While GroenLinks-PvdA had many mentions of the issue nature and environment in their political agenda, there is no correlation with the media agenda of NRC. GroenLinks-PvdA, VVD and NSC did not have any significant correlations with the NRC, suggesting that factors beyond party campaigning may relate to the media agenda of NRC. 47 The correlations of the political agendas with the media agenda of De Telegraaf reveal a similar outcome. PVV shows a significant positive correlation between the issue nature and environment and De Telegraaf. Revealing an alignment between the political agenda of the PVV and the media agenda of De Telegraaf. The other issues mentioned by PVV, and all of the issues mentioned by GroenLinks-PvdA, VVD and NSC did not have significant correlations with the media agenda of De Telegraaf. Similar to the media agenda of NRC, suggesting that factors beyond party campaigning relate to the media agenda of De Telegraaf. The absence of significant correlations for GroenLinks-PvdA and VVD and the NSC across both Facebook posts and advertisements underscores the complexity of the media landscape, where the relationship between the media and political agendas is not always straightforward. When looking at the role of professionalisation, the Facebook content of GroenLinks-PvdA has the highest degree of professionalisation in their campaign. However, the party does not share the same agenda with any of the newspapers. These findings imply that despite the parties’ professionalisation efforts, they did not have an agenda similar to those of the media, or their focal issues did not align with what the newspapers deemed newsworthy during the election period. The same can be said for the VVD, which shows many indicators of professionalisation, and NSC, which shows very few indicators of professionalisation. The outlier in these results is the PVV, which shows strong professionalisation indicators on call- to-action and engagement and has significant positive correlations with both newspapers on two issues. This finding suggests that PVV's professionalised Facebook campaign content effectively mirrors the media's agenda or vice versa, specifically on nature and environment and immigration. This provides insights into the role that professionalisation has on the interaction between the media and political agendas. It is not possible to present definitive conclusions, the results suggest that the professionalisation of parties may play a role in synchronising their agendas with that of the media; however, it is not very straightforward or the sole determinant. Media and political agenda alignment could be more pronounced for parties with clear ownership of specific issues. Professionalisation can, in some cases, increase the number of times their ‘owned’ issues are mentioned in their political agendas as well as that of the media, in line with the issue ownership theory. For example, the PVV has strong attention to the issue of immigration in their posts. At the same time, newspapers also mention this issue many times. 48 5. Conclusion and Discussion This thesis has examined the interaction between the media agendas of De Telegraaf and NRC and the political agendas through the Facebook campaigns of PVV, GroenLinks-PvdA, VVD and NSC during the final two months leading up to the 2023 Dutch national elections. The study aimed to answer the central research question: What is the relationship between the media agendas of De Telegraaf and NRC and the political agendas of PVV, GroenLinks-PvdA, VVD and NSC in the final two months of the 2023 Dutch national elections, and what relationship do these agendas have with the (professionalised) political campaigns on Facebook? This study analysed how these newspapers covered issues and political leaders through agenda-setting and considered the role of professionalisation in this relationship. 5.1. Conclusion The Media and Political Agendas The analysis revealed different approaches in the media agendas between a tabloid, De Telegraaf, and a broadsheet, NRC. Next to societal issues, such as nature and environment and immigration, De Telegraaf tended to cover immediate and tangible issues such as taxes, public transport, and the Israel-Palestine conflict, prioritising current events impacting daily life. In contrast, NRC had a stronger emphasis on long-term societal challenges, notably nature and environment and education and science. This focus aligns with the work of Bastos (2019), who argues that broadsheets have a more thorough, possibly investigative, approach, and tabloids showcase current issues. Regarding the four political leaders analysed in this thesis, Pieter Omtzigt and Frans Timmermans received significant attention across both newspapers, suggesting their central roles in the political discourse at the time, possibly due to their active involvement in key issues or their positions within their respective parties. The political agendas of the parties varied in their focus: GroenLinks-PvdA consistently focused on the issue of nature and environment, aligning with their established issue ownership as van der Meer (2022) stated. PVV highlighted immigration and the Israel- Palestine conflict, reinforcing their ownership of these issues as specified by Kanne & van der Schelde (2023). VVD’s political agenda is characterised by a notable absence of specific issue discussions, focusing instead on a highly personalised strategy centred around their leader, diverging from the expected issue ownership on the economy as described by van der Meer (2022). NSC also lacked a clear issue focus, reflecting a generalised approach in their 49 campaign, which is in line with research by Kanne & van der Schelde (2023). Their content did show an emphasis on visual engagement over textual elaboration. Professionalised Campaigns on Facebook The examination of the six indicators of professional campaigns on Facebook, as outlined by Heiss et al. (2018), Lilleker (2019), Römmele & Gibson (2020), and Jacobs & Spierings (2018), revealed variations in how parties utilised these indicators. GroenLinks-PvdA displayed the most comprehensive approach, effectively incorporating calls-to-action, significant advertisement spending, and a balanced demographic targeting strategy. VVD employed all indicators but showed limited use of calls-to-action, potentially neglecting mobilisation opportunities. Interestingly, alongside GroenLinks-PvdA, VVD was the only party to utilise Facebook advertisements. In their Facebook posts, PVV demonstrated a strong focus on calls-to-action and achieved high engagement, suggesting a strategic approach to digital campaigning. However, their lack of advertisements and limited demographic targeting suggest a less refined strategy compared to GroenLinks-PvdA and VVD. Finally, NSC exhibited the lowest level of professionalisation with no advertisements, minimal call-to- action practices, and the lowest engagement numbers. Relationship Between The Media and Political Agendas The relationship between the political agendas on Facebook and the media agendas of De Telegraaf and NRC proved to be complex, aligning with the notion of these complex dynamics by Gilardi et al. (2022). PVV, which demonstrated strong indicators of professionalisation in their campaign, showed significant positive correlations in both newspapers on the issues of nature and environment and immigration. This suggests potential alignment with their key issues. This alignment may be due to a combination of PVV’s professionalised content on Facebook and existing media interest in these issues. GroenLinks-PvdA’s focus on nature and environment was frequently mentioned in media agendas, but despite their highly professionalised campaign, there was no significant correlation between their political agenda and media agendas. This finding highlights that even well-executed professional campaigns may not always align with the media agendas. The analysis of VVD and NSC also resulted in no significant correlations between their 50 political agendas and the media agendas. The lack of correlations could be due to the unpredictable nature of media dynamics or the influence of external factors, such as institutional factors, timing and content characteristics as suggested by Vliegenthart & Walgrave (2019) and Strömback & Kiousis (2014). The Role of Professionalisation in the Media and Political Agenda Relationship The findings do not reveal a unidirectional outcome on the relationship between the political and media agendas and no definitive outcome or role for professionalisation in the relation between the political and media agendas. While professionalisation could increase the frequency of issue mentions in the political and media agendas, as seen with PVV’s focus on immigration and their significant correlations with the media agendas, this is not uniform for all parties. An explanation could lie in the differences in the tools used within these professional campaigns. Referring back to Mykkänen et al. (2022), populist and extremist parties, like PVV, may use more emotional elements in their professionalised campaign tools. This contrasts with parties that may have used a more rational-oriented toolbox such as GroenLinks-PvdA, VVD and NSC. The findings for these three parties, particularly GroenLinks-PvdA, with their emphasis and issue ownership on nature and environment and having a high degree of professionalisation, did not result in significant correlations between their political agenda and the media agendas. This lack of correlation highlights the complexity of professionalisation in the relationship between the political and media agendas. The professionalisation of political campaigns can be a tool for parties to set their agendas or align with the media agendas, but its effectiveness is contingent on various other (external) factors, such as the tools used and the pre-existing media interest in certain issues. 51 5.2. Discussion This study examined the dynamics between media agendas and political agendas during the final two months of the 2023 Dutch national elections, and the role of professionalisation. The findings indicated differences in the issues covered by De Telegraaf and NRC, reflecting their editorial focus. Political leaders like Pieter Omtzigt and Frans Timmermans were prominently featured, underscoring their importance in the political discourse. The political agendas of major parties revealed issue ownership for some, such as GroenLinks-PvdA with nature and environment, and PVV with immigration and the Israel-Palestine conflict. In contrast, VVD and NSC displayed a less clear issue focus, instead opting for a more personalised or generalised campaign approach. This variation highlights the strategic differences among parties in addressing their strengths and audience interests. Reflecting on the findings, the relationship between political and media agendas is complex and varies among parties. This complexity aligns with existing research that suggests numerous external factors in the relationship between the media and political agendas. However, the consistent emphasis by GroenLinks-PvdA and PVV on their key issues (nature and environment and immigration) suggests that political parties which strategically highlight issues they are perceived to own can strengthen their position on those issues. Additionally, the key issues of GroenLinks-PvdA and PVV are also mentioned most by both newspapers. This could mean that the media is specifically interested in the owned issues by GroenLinks- PvdA and PVV or the parties focused on the issues most mentioned by the media. Other reasons could be that the media had previous interest in the issues or that these issues dominated the public discourse in other ways. This research does not have a conclusive answer to this; with the PVV’s political agenda having significant correlations with both newspapers and GroenLinks-PvdA, VVD and NSC not showing any significant correlations, there is no definitive outcome. The findings for the role of professionalisation are also not definitive. The parties that show a high degree of professionalisation, such as GroenLinks-PvdA and VVD, did not reveal any significant correlations between their political agendas and the media agendas, as well as NSC, with low professionalisation. PVV, with strong professionalisation indicators, especially on mobilisation (call-to-action) and engagement (likes, comments and shares), does have significant correlations. This could mean that these indicators play an important part in the role that professionalisation plays. However, the influence of external factors must be considered. Research builds on the difference in tools used, with PVV possibly having more 52 emotional elements in their professionalised campaign. This is a reasonable explanation because of the high engagement numbers of the PVV. In this case, the professionalisation aspect of political campaigns can be a relevant tool for political parties to make strategic choices in choosing their key issues and increasing issue ownership, but a uniform role of professionalisation among parties is not found. The findings of this study have contributed to an understanding of the relationship between the media and political agendas and the role that professionalisation plays in contemporary Dutch elections. While the findings show a complex landscape, for political campaign strategists, these findings highlight the many possible strategies to engage with voters and the need to consider the efficacy of professionalisation on the agenda-setting possibilities of political campaigns. For media practitioners, understanding media agenda-setting influence can guide more informed and responsible coverage of political issues, ensuring a more balanced representation of diverse political agendas. Limitations and Future Research The study’s methodology, while comprehensive, has limitations that need consideration. The Facebook content that was included in this thesis did not feature a wide range of issues, which restricted the ability to comprehensively assess the relationship between the political and media agendas. Additionally, the focus on just two newspapers may have provided a partial picture of the media landscape, potentially limiting the generalisability of the findings. The relatively small volume of posts and lack of advertisements by PVV and NSC, coupled with differences in the amount and diversity of issues among the political parties, may have reduced the statistical robustness of the conclusions and introduced biases toward more active parties. Despite these limitations, conclusions have been drawn for each party separately, offering relevant insights into their individual strategies and media interactions. Given these dynamics, there is a clear need for further investigation into how the political and media agenda relationship develops over time and across different electoral contexts and social media platforms, especially considering the role of professionalisation. Future research should consider a more extended timeframe to capture the evolution of media and political agendas over the entire election cycle. Including a broader range of social media platforms and offline professionalised campaign strategies which could provide a more comprehensive understanding of political communication. Additionally, examining the role of media 53 ownership and editorial policies could uncover deeper insights into the factors influencing media agendas. Comparative studies across different countries and election contexts could also help generalise the findings and enhance the understanding of media-political relationships in diverse settings. Furthermore, various sentiment analysis methods can be pursued to understand the tone of voice regarding the articles and posts, such as crowd-coding practices and applying machine learning for analysis. 54 References Arceneaux, P., Borden, J., & Golan, G. (2019). The News Management Function of Political Public Relations. In J. Strömback, & S. Kiousis, Political Public Relations: Concepts, Principles, and Applications (pp. 126-145). New York: Routledge. Banda, K. K. (2013). The Dynamics of Campaign Issue Agendas. State Politics & Policy Quarterly 13:4, 446-470. Barberá, P., Casas, A., Nagler, J., Egan, P., Bonneau, R., Jost, J., & Tucker, J. A. (2019). Who Leads? Who Follows? Measuring Issue Attention and Agenda Setting by Legislators and the Mass Public Using Social Media Data. American Political Science Review 113:4, 883-901. Bartholomé, G., Lecheler, S., & de Vreese, C. (2018). Towards a typology of conflict frames: Substantiveness and interventionism in political conflict news. Journalism Studies 19:12, 1689-1711. Bastos, M. (2019). Tabloid Journalism. In T. Vos, F. Hanusch, M. Geertsema-Sligh, A. Sehl, & D. Dimitrakopoulou, The International Encyclopedia of Journalism Studies (pp. 1-6). Wiley. Beraldo, D., Milan, S., de Vos, J., Agosti, C., Sotic, B. N., Vliegenthart, R., . . . Votta, F. (2021, 03 11). Political advertising exposed: tracking Facebook ads in the 2021 Dutch elections. Retrieved from Internet Policy Review: https://policyreview.info/articles/news/political-advertising-exposed-tracking- facebook-ads-2021-dutch-elections/1543 Camp, K. V. (2017). Issue ownership as a determinant of political parties’ media coverage. Communication 43:1, 1-21. Chadwick, A. (2017). The Hybrid Media System. New York: Oxford University Press. Chu, X., Vliegenthart, R., Otto, L., Lecheler, S., de Vreese, C., & Kruikemeier, S. (2024). Do Online Ads Sway Voters? Understanding the Persuasiveness of Online Political Ads. Political Communication 41:2, 290-314. Cole, R. (2023). Inter-Rater Reliability Methods in Qualitative Case Study Research. Sociological Methods & Research, 1-32. 55 Commissariaat voor de Media. (2023). Mediamonitor 2023. Commissariaat voor de Media. de Fortuny, E., De Smedt, T., Martens, D., & Daelemans, W. (2012). Media coverage in times of political crisis: A text mining approach. Expert Systems with Applications 39, 11616-11622. de Jong, C., & Theulings, R. (2023, 09 14). Bestaanszekerheid belangrijkste verkiezingsthema? Zo wordt bepaald waar het over gaat tijdens de campagne. Retrieved from EenVandaag: https://eenvandaag.avrotros.nl/item/bestaanszekerheid- belangrijkste-verkiezingsthema-zo-wordt-bepaald-waar-het-over-gaat-tijdens-de- campagne/ de Ridder, J., Vliegenthart, R., & Zuure, J. (2020). Doen, durven of de waarheid? Democratie in digitale tijden. Amsterdam: Amsterdam University Press. De Smedt, T., & Daelemans, W. (2012). “Vreselijk mooi!” (terribly beautiful): A Subjectivity Lexicon for Dutch Adjectives. Proceedings of the 8th international conference on Language Resources and Evaluation (LREC2012) (pp. 3568-3572). European Language Resources Association (ELRA). Dege, D., & Brüggeman, P. (2023). Marketing analytics with RStudio: a software review. Journal of Marketing Analytics , 1-6. Druckman, J. N. (2004). Priming the Vote: Campaign Effects in a U.S. Senate Election. Political Psychology 25:4, 577-594. Dutch Parliamentary Election Survey. (2017). Statistical Analysis. Retrieved from Stichting Kiezersonderzoek Nederland: https://www.dpes.nl/statistical-analysis/ Enli, G. (2017). Twitter as arena for the authentic outsider: exploring the social media campaigns of Trump and Clinton in the 2016 US presidential election. European Journal of Communication 32:1, 50-61. Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm . Journal of Communication 43:4, 51-58. Esser, F., & Neuberger, C. (2019). Realizing the democratic functions of journalism in the digital age: New alliances and a return to old values. Journalism 20:1, 194-197. 56 Esser, F., & Strömback, J. (2014). Mediatization of Politics: Understanding the Transformation of Western Democracies. Palgrave Macmillan. Farkas, X., & Bene, M. (2021). Images, Politicians, and Social Media: Patterns and Effects of Policians' Image-Based Political Communication Strategies on Social Media. The International Journal of Press/Politics 26:1, 119-142. Feezell, J. T. (2018). Agenda Setting through Social Media: The Importance of Incidental News Exposure and Social Filtering in the Digital Era. Political Research Quarterly 71:2, 482-494. Geers, S., & Bos, L. (2017). Priming Issues, Party Visibility, and PartyEvaluations: The Impact on Vote Switching. Political Communication 34, 344–366. Gibson, R. K., & Römmele, A. (2009). Measuring the professionalization of political campaigning. Party Politics 15:3, 265-293. Gibson, R., & Römmele, A. (2001). A Party-Centered Theory of Professionalized Campaigning. Press/Politics 6:4, 31-43. Gilardi, F., Gessler, T., Kubli, M., & Müller, S. (2022). Social Media and Political Agenda Setting. Political Communication 39:1, 39-60. Green, J., & Jennings, W. (2011). The dynamics of issue competence and vote for parties in and out of power: An analysis of valence in Britain, 1979–1997ejpr_2004 1..35. European Journal of Political Research 51:4, 1-35. Green, M. C. (2004). Transportation Into Narrative Worlds: The Role of Prior Knowledge and Perceived Realism. Discourse Processes, 38:2, 247-266. Harder, R. A., Sevenans, J., & Van Aelst, P. (2017). Intermedia Agenda Setting in the Social Media Age: How Traditional Players Dominate the News Agenda in Election Times. The International Journal of Press/Politics 22:2, 1-19. Heiss, R., Schmuck, D., & Matthes, J. (2018). https://doi.org/10.1080/1369118X.2018.1445273. What drives interaction in political actors’ Facebook posts? Profile and content predictors of user engagement and political actors’ reactions, 1-17. 57 Hoeven, R. v. (2012). Het succes van een kwaliteitskrant : de ontstaansgeschiedenis van NRC Handelsblad [Dissertation thesis]. Leiden: Universiteit Leiden. Hoof, A. v. (2000). Kranten met karakter: identiteit van kranten vanuit kwaliteitsperspectief [Doctoral dissertation, Radboud University]. Amsterdam: Thela Thesis. Ipsos. (2023). Tussenmeting Tweede Kamerverkiezingen 2023. Amsterdam: Ipsos. Jacobs, K., Lubber, M., Spierings, N., & Voogd, R. (2024, 03 27). Kiezersonderzoek 2023, deel 2: GroenLinks-PvdA. Retrieved from Stuk rood vlees: https://stukroodvlees.nl/kiezersonderzoek-2023-deel-2-groenlinks-pvda/ Kanne, P., & van der Schelde, A. (2023, 11 18). Strategische stem kan het verschil maken. Retrieved from Ipsos I&O: https://www.ioresearch.nl/actueel/strategische-stem-kan- het-verschil-maken/ Kiesraad. (2023, 12 01). Kiesraad stelt uitslag Tweede Kamerverkiezing 22 november 2023 vast. Retrieved from Kiesraad: https://www.kiesraad.nl/actueel/nieuws/2023/12/01/kiesraad-stelt-uitslag-tweede- kamerverkiezing-22-november-2023-vast Kleinnijenhuis, J., van Hoof, A. M., Oegema, D., & de Ridder, J. A. (2007). A Test of Rivaling Approaches to Explain News Effects: News on Issue Positions of Parties, Real-World Developments, Support and Criticism, and Success and Failure. Journal of Communication 57, 366-384. Kreiss, D. (2016). The Dynamics of Technology-Intensive Campaigning. In D. Kreiss, Prototype Politics: Technology-Intensive Campaigning and the Data of Democracy (pp. 204-220). Oxford University Press. Kruikemeier, S., Sezgin, M., & Boerman, S. C. (2016). Political Microtargeting: Relationship Between Personalized Advertising on Facebook and Voters’ Responses. Cyberpsychology, Behavior, and Social Networks 19:6, 367-372. Lees-Marshment, J. (2001). The Marriage of Politics and Marketing. Political Studies 49, 692-713. Lilleker, D. (2020). Political Public Relations and Election Campaigning. In J. Strömback, & S. Kiousis, Political Public Relations. Concepts, Principles, and Applications (pp. 187-207). Routledge: New York. 58 Lilleker, D. G. (2019). The Power of Visual Political Communication: Pictorial Politics Through the Lens of Communication Psychology. In D. G. Lilleker, A. Veneti, D. Jackson, & D. G. Lilleker, Visual Political Communication (pp. 37-51). Palgrave Macmillan Cham. Lilleker, D. G., & Negrine, R. (2002). Professionalization: Of What? Since When? By Whom? Press/Politics, 98-103. Magin, M., Podschuweit, N., Haßler, J., & Rußmann, U. (2017). Campaigning in the Fourth Age of Political Communication. A Multi-Method Study on the Use of Facebook by German and Austrian Parties in the 2013 National Election Campaigns. Information, Communication & Society 29:11, 1698-1719. McCombs, M. (2005). A Look at Agenda-setting: past, present and future. Journalism Studies 6:4, 543-557. McCombs, M. (2014). Setting the Agenda: The Mass Media and Public Opinion. Cambridge: Polity Press. McCombs, M. E., & Shaw, D. L. (1972). The Agenda-Setting Function of Mass Media. The Public Opinion Quarterly 36:2, 176-187. McLaughlin, B., & Velez, J. A. (2019). Imagined Politics: How Different Media Platforms Transport Citizens Into Political Narratives. Social Science Computer Review, 37:1, 22-37. Meijer, M. M., & Keinnijenhuis, J. (2006). Issue news and corporate reputation: Applying the theories of agenda setting and issue ownership in the field of business communication. Journal of Communication 56:3, 543-559. Meiserli, F. (2021). On Political Communication: Is Social Media More Effective? Resolusi 4:2, 71-78. Meyer, T. M., Haselmayer, M., & Wagner, M. (2017). Who Gets into the Papers? Party Campaign Messages and the Media. Britisch Journal of Political Science 50:1, 281- 302. Mykkänen, J., Nord, L., & Moring, T. (2022). Ten years after: Is the party-centered theory of campaign professionalization still valid? Party Politics 28:6, 1176-1186. 59 NOS. (2023, 07 07). Kabinet gevallen, geen akkoord over asielmaatregelen. Retrieved from NOS: https://nos.nl/collectie/13942/artikel/2481938-kabinet-gevallen-geen-akkoord- over-asielmaatregelen Panagopoulos, C. (2006). Political Consultants, Campaign Professionalization, and Media Attention. Political Science & Politics 39:4, 867-869. Partheymüller, J. (2017). Agenda-Setting Dynamics during the Campaign Period. In H. Schoen, S. Roßteutscher, R. Schmitt-Beck, B. Weßels, & C. Wolf, Voters and Voting in Context: Multiple Contexts and the Heterogeneous German Electorate (pp. 12-41). Rose, S., Sprinks, N., & Canhoto, A. I. (2015). Quantitative content analysis. In S. Rose, N. Sprinks, & A. I. Canhoto, Management Research: Applying the Principles (pp. 1-9). Routledge. Ruthardt, C. (2023). Marketizing Elections? An Analysis of Audience Segmentation and Targeted Messaging as Media Strategy in German Electoral Campaigning [Master's Thesis, SciencesPo]. Retrieved from https://www.sciencespo.fr/ecole-affaires- publiques/sites/sciencespo.fr.ecole-affaires-publiques/files/RUTHARDT.pdf Scheffauer, R., Goyanes, M., & Gil de Zuñiga, H. (2023). Social media algorithmic versus professional journalists’ news selection: Effects of gate keeping on traditional and social media news trust. Journalism, 1-24. Schyns, V. (2023, 11 23). Jubelende Wilders domineert de voorpagina’s van Nederlandse kranten. Retrieved from NRC: https://www.nrc.nl/nieuws/2023/11/23/jubelende- wilders-domineert-de-voorpaginas-van-nederlandse-kranten-a4181998 Sens, A. (2009). Het geheim van De Telegraaf: geschiedenis van een krant [Review]. Persmuseum Amsterdam, 144-145. Shahanan, E. A., McBeth, M. K., & Hathaway, P. L. (2011). Narrative Policy Framework: The Influence of Media Policy Narratives on Public Opinion. Politics & Policy 39:3, 373- 400. Shenhav, S. R. (2006). Political Narratives and Political Reality. International Political Science Review 27:3, 245-262. Sipma, T., & Berning, C. C. (2021). Economic conditions and populist radical right voting: The role of issue salience. Electoral Studies 74, 1-10. 60 Sorensen, L. (2018). Populist Communication in the New Media Environment: A Cross- Regional Comparative Perspective. Palgrave Communications 4:1, 1-12. Spierings, N., & Jacobs, K. (2019). Political parties and social media campaigning. Acta Polit 54, 145-173. Springer, S. A., & Harwood, J. (2015). The Influence of Episodic and Thematic Frames on Policy and Group Attitudes: Mediational Analysis. Human Communication Research 41:2, 226-244. Strömback, J., & Dimitrova, D. V. (2011). Mediatization and Media Interventionism: A Comparative Analysis of Sweden and the United States. The International Journal of Press/Politics 16:1, 30-49. Strömback, J., & Kiousis, S. (2014). Strategic political communication in election campaigns. In C. Reinemann, Political Communication (pp. 109-128). Berlin: De Gruyter Mouton. Stromer-Galley, J. (2019). 2008: Networked Campaigning and Controlled Interactivity. In J. Stromer-Galley, Presidential Campaigning in the Internet Age (pp. 110-146). Oxford University Press. Takens, J. (2013). Media Logic in Election Campaign Coverage. In J. H. Takens, Media Logic and Electoral Democracy. [PhD-Thesis - Research and graduation internal; Vrije Universiteit Amsterdam] (pp. 79-98). Amsterdam: Box Press. Takens, J., van Atteveldt, W., van Hoof, A., & Kleinnijenhuis, J. (2013). Media logic in election campaign coverage. European Journal of Communication 28:3, 277-293. The Amsterdam School of Communications Research. (2007). Inhoudsanalyse voor de campagne van de Tweede Kamerverkiezingen van 22 november 2006. Amsterdam: Universiteit van Amsterdam. Thesen, G. (2013). When good news is scarce and bad news is good: Government responsibilities and opposition possibilities in political agenda-setting. European Journal of Political Research 52, 364-389. Van Aelst, P., & Walgrave, S. (2017). Information and Arena: The Dual Function of the News Media for Political Elites. In P. Van Aelst, & S. Walgrave, How Political Actors Use the Media (pp. 1-20). London: Palgrave. 61 Van Aelst, P., Scheafer, T., & Stanyer, J. (2011). The personalization of mediated political communication: A review of concepts, operationalizations and key findings. Journalism 13:2, 203-220. van Atteveldt, W., van der Velden, M. A., & Boukes , M. (2021). The Validity of Sentiment Analysis: Comparing Manual Annotation, Crowd-Coding, Dictionary Approaches, and Machine Learning Algorithms. Communication Methods and Measures 15:2, 121-140. van der Meer, T. (2022). Associative issue ownership in a highly fragmented multiparty context: The Netherlands (2021). Acta Politica, 1-21. van der Pas, D. J., van der Brug, W., & Vliegenthart, R. (2017). Political Parallelism in Media and Political Agenda-Setting. Political Communication 34:4, 491-510. Vermeer, S., Remoortere, V. A., & Vliegenthart, R. (2022). Still going strong? The role of traditional media in the 2021 Dutch parliamentary elections. Acta Politica, 1-17. Vliegenthart, R. (2012). The Professionalization of Political Communication? A Longitudinal Analysis of Dutch Election Campaign Posters. American Behavioral Scientist 56:2, 135-150. Vliegenthart, R., & Walgrave, S. (2019). The Media Agenda. In F. R. Baumgartner, C. Breunig, & E. Grossman, Comparative Policy Agendas: Theory, Tools, Data (pp. 271- 281). Oxford University Press. Vliegenthart, R., Boomhaarden, H. G., Van Aelst, P., & de Vreese, C. (2010). Covering the US presidential election in Western Europe: A cross-national comparison. Acta Politica 45, 444-467. Walgrave , S., & Van Aelst, P. (2006). The Contingency of the Mass Media’s Political Agenda Setting Power: Toward a Preliminary Theory. Journal of Communication 56, 88-109. Weaver, D. H. (2007). Thoughts on Agenda Setting, Framing, and Priming. Journal of Communication 57, 142-147. Zuiderveen Borgesius, F. J., Möller, J., Kruikemeier, S., Ó Fathaigh, R., Irion, K., Dobber, T., . . . de Vreese, C. (2018). Online Political Microtargeting: Promises and Threats for Democracy. Utrecht Law Review 14:1, 82-96. 62 Appendix 1 – Newspaper Codebook A. General information: Note. All questions are coded and analysed through statistical computing software R. AA. Item number A1. From which newspaper is the item sourced? 1. De Telegraaf 2. NRC A2. What is the publication date of the story? Code in DDMMYY format B. Campaign news filter Any message in an item that meets one of the following criteria: - National party leader, regardless of whether they act in the role of party leader, - Events in the context of the elections, - Reference to elections, election programs, or campaigns, or it concerns the (future) cabinet. - (Current/future): about the composition of policy. This also includes the minister's policy. These criteria are specified for coding purposes using the following reference words. Items containing these words are considered campaign news: ‘verkiezingen / elections’, ‘verkiezing / election’, ‘Tweede Kamer-verkiezingen / elections’, ‘verkiezingsdag / election day’, ‘verkiezingsprogramma / election manifesto’, ‘partijprogramma / election manifesto , ‘campagne / campaign’, ‘peilingen / polls’, ‘peiling / poll’, ‘zetels / seats (for Parliament)’, ‘zetel / seat (for Parliament)’, ‘regeren / govern’, ‘verkiezingscampagne / election campaign’,’kandidaat / candidate’, ‘kandidaten / candidates’, ‘nieuw kabinet / new cabinet’ stemmen / voting’, stemgedrag / voting behaviour’, ‘regeringsdeelname / government participation’, ‘kiezers / voters’, ‘nieuwe premier / new minister-president’, ‘nieuwe minister- president / new minister-president’, ‘lijsttrekker / party leader’, ‘verkiezingsdebat / election debate’ . 63 B1. Does the item concern campaign news? 1. Yes 2. No Note. If the answer to question B1 is no, the item is not included in the dataset. C. Selection of newspaper items C1. What is the headline of the article? C2. Page number C3. How large is the item in words? C4. What type of item is it? Note. Items within NRC and De Telegraaf are combined in relevant grouped sections. The relevant sections within the newspaper are included in brackets for coding purposes) 1. News report / interview / background story (NRC: Binnenland, economie, in het nieuws, voorpagina, weekend, wetenschap) (De Telegraaf: Binnenland, financieel, journaal, regio, reportage) 2. Column / Opinion piece / piece on the opinion page (NRC: Achterpagina, opinie, opinie & debat) (De Telegraaf: Wat u zegt) 3. Other (NRC: Boeken, cultuur, sport) (De Telegraaf: Sport) D. Issues Note. For coding purposes, relevant and connected Dutch words are mentioned in brackets and the English translation added. D1. What issues can be found in the item? 1. Foreign policy 64 Coded sub-categories: a. General foreign policy (buitenlands beleid / foreign policy, internationale relaties / international relations) b. European Union (Europese Unie / European Union, Europa / Europe, EU- beleid / EU-policy) c. Defense (defensie / defense, leger / army, militaire samenwerking / military cooperation) d. Development aid (ontwikkelingssamenwerking / development aid) e. Israeli-Palestine conflict (Israel / Israel, Palestina / Palestine, Gaza / Gaza, Westelijke Jordaanoever / West Bank) 2. Safety Coded sub-categories: a. Safety / crime (veiligheid / safety, criminaliteit / crime) b. Judiciary (justitie / judiciary) c. Terrorism domestically (terrorismebestrijding / counterterrorism) 3. Immigration and integration Coded sub-categories: a. Integration of minorities (integratie / integration, multiculturele samenleving / multicultural society) b. Immigration (immigratie / immigration, asiel beleid / asylum policy, vluchteling / refugee, gelukzoekers / fortune seekers, massaimmigratie / mass immigration) 4. The democratic system Coded sub-categories: a. Participation (burgerparticipatie / citizen participation, participatie / participation, burgerinitiatief / citizen initiative, vertrouwen / trust), b. Rule of law (constitutioneel hof / constitutional court, rechtsstaat / the rule of law) 5. Economy Coded sub-categories: a. General economic policy and economic stimulation (economie | economy, economisch beleid / economic policy, groei / growth) b. Government finances / national debt / budget deficit (staatsschuld / national debt, begroting / budget, financiën / finances) 65 c. Inflation / price increases / monetary policy / interest rates (inflatie / inflation, rente / rates, lastenverlichting / monetary relief) d. Privatization / liberalisation / free market (privatisering / privatization, marktwerking / market competition, vrije markt / free market) e. Taxation (belasting / tax, fiscal beleid / taxation policy) 6. Social Affairs Coded sub-categories: a. General social policy (social beleid / social policy, bestaanszekerheid / social security) b. Employment / labour market (banen / jobs, arbeidsmarkt / labour market) c. Wages and collective labour agreements (loon / wages, CAO / collective labour agreements) d. Social division / poverty in the Netherlands (armoede / poverty, ongelijkheid / inequality) e. Benefits / Social assistance (bijstand / benefits, sociale zekerheid / social assistance) f. Pensions and aging policy (pensioen / pensions, ouderenbeleid / aging policy) g. Youth policy (jeugd / youth, jongerenbeleid / youth policy, leenstelsel / student loan program, pechgeneratie / bad luck generations (students)) h. Childcare (kinderopvang / childcare) i. Emancipation, for example of women, homosexuals, or immigrants (emancipatie / emancipation, gelijkheid / equality) 7. Health and Sport Coded sub-categories: a. Healthcare (gezondheid / health, gezondheidszorg / health care, zorgverzekering / health insurance) b. Ethical issues (ethiek / ethics, morele vraagstukken / ethical issues) c. Drug policy (drugs / drugs, verslavingszorg / addiction care, wiet / weed) d. Sport (sport / sport, beweging / exercise) 8. Nature, Environment, and Agriculture Coded sub-categories: a. Nature and environment (milieu / environment, natuur / nature, natuurbehoud / nature conservation, klimaatverandering / climate change, klimaat / climate, fossiel / fossil) 66 b. Nitrogen crisis (stikstof / nitrogen) c. Animals (dierenwelzijn / animal welfare, dierenrechten / animal rights) d. Agriculture and horticulture and food safety (landbouw / agriculture, voedselproductie / food production, voedselveiligheid / food safety, tuinbouw / horticulture, boeren / farmers) 9. Transportation Coded sub-categories: a. Infrastructure (infrastructuur / infrastructure) b. Road traffic (verkeer / traffic, wegen / roads) c. Schiphol / aviation (Schiphol / Schiphol Amsterdam Airport, luchtvaart / aviation, vliegveld Lelystad / Lelystad Airport) d. Public transport (openbaar vervoer / public transportation, treinen / trains, bussen / busses) 10. Housing Coded sub-category: a. (Public) housing and housing market(wonen / living, sociale woningbouw / social housing, hypotheekrente / mortgage interest, woningmarkt / housing market, huizenmarkt / housing market) 11. Education and Culture Coded sub-categories: a. Education and science (onderwijs / education, school / school, wetenschap / science) b. Art and culture (cultuur / culture, kunst / art) E. Political leaders E1. Which political leader is/are mentioned in the item? 1. PVV: Geert Wilders 2. VVD: Dilan Yeşilgöz-Zegerius 3. NSC: Pieter Omtzigt 4. GroenLinks-PvdA: Frans Timmermans 99. None 67 Appendix 2 – Facebook Content Codebook A. General information Note. All questions are coded and analysed through statistical computing software R. AA. Item number A1. From which party is the item sourced? 1. PVV 2. GroenLinks-PvdA 3. VVD 4. NSC A2. What is the publication date of the item? Code in DDMMYY format A3. How large is the item in words? A4. Content type 1. Organic post 2. Advertisement A5. Social media typology 1. Attachment (such as a link, with or without text) 2. Video (with out without text) 3. Text (without images) 4. Image (without text) 5. Carousel (combination of video/image/text) B. Issues Note. Relevant and connected Dutch words are mentioned in brackets for coding purposes, and the English translation is added. B1. What issues can be found in the item? 1. Foreign policy 68 Coded sub-categories: a. General foreign policy (buitenlands beleid / foreign policy, internationale relaties / international relations) b. European Union (Europese Unie / European Union, Europa / Europe, EU- beleid / EU-policy) c. Defense (defensie / defense, leger / army, militaire samenwerking / military cooperation) d. Development aid (ontwikkelingssamenwerking / development aid) e. Israeli-Palestine conflict (Israel / Israel, Palestina / Palestine, Gaza / Gaza, Westelijke Jordaanoever / West Bank) 2. Safety Coded sub-categories: a. Safety / crime (veiligheid / safety, criminaliteit / crime) b. Judiciary (justitie / judiciary) c. Terrorism domestically (terrorismebestrijding / counterterrorism) 3. Immigration and integration Coded sub-categories: a. Integration of minorities (integratie / integration, multiculturele samenleving / multicultural society) b. Immigration (immigratie / immigration, asiel beleid / asylum policy, vluchteling / refugee, gelukzoekers / fortune seekers, massaimmigratie / mass immigration) 4. The democratic system Coded sub-categories: a. Participation (burgerparticipatie / citizen participation, participatie / participation, burgerinitiatief / citizen initiative, vertrouwen / trust), b. Rule of law (constitutioneel hof / constitutional court, rechtsstaat / the rule of law) 5. Economy Coded sub-categories: a. General economic policy and economic stimulation (economie | economy, economisch beleid / economic policy, groei / growth) b. Government finances / national debt / budget deficit (staatsschuld / national debt, begroting / budget, financiën / finances) 69 c. Inflation / price increases / monetary policy / interest rates (inflatie / inflation, rente / rates, lastenverlichting / monetary relief) d. Privatization / liberalization / free market (privatisering / privatisation, marktwerking / market competition, vrije markt / free market) e. Taxation (belasting / tax, fiscal beleid / taxation policy) 6. Social Affairs Coded sub-categories: a. General social policy (social beleid / social policy, bestaanszekerheid / social security) b. Employment / labour market (banen / jobs, arbeidsmarkt / labour market) c. Wages and collective labour agreements (loon / wages, CAO / collective labour agreements) d. Social division / poverty in the Netherlands (armoede / poverty, ongelijkheid / inequality) e. Benefits / Social assistance (bijstand / benefits, sociale zekerheid / social assistance) f. Pensions and aging policy (pensioen / pensions, ouderenbeleid / aging policy) g. Youth policy (jeugd / youth, jongerenbeleid / youth policy, leenstelsel / student loan program, pechgeneratie / bad luck generations (students)) h. Childcare (kinderopvang / childcare) i. Emancipation, for example of women, homosexuals, or immigrants (emancipatie / emancipation, gelijkheid / equality) 7. Health and Sport Coded sub-categories: a. Healthcare (gezondheid / health, gezondheidszorg / health care, zorgverzekering / health insurance) b. Ethical issues (ethiek / ethics, morele vraagstukken / ethical issues) c. Drug policy (drugs / drugs, verslavingszorg / addiction care, wiet / weed) d. Sport (sport / sport, beweging / exercise) 8. Nature, Environment, and Agriculture Coded sub-categories: a. Nature and environment (milieu / environment, natuur / nature, natuurbehoud / nature conservation, klimaatverandering / climate change, klimaat / climate, fossiel / fossil) 70 b. Nitrogen crisis (stikstof / nitrogen) c. Animals (dierenwelzijn / animal welfare, dierenrechten / animal rights) d. Agriculture and horticulture and food safety (landbouw / agriculture, voedselproductie / food production, voedselveiligheid / food safety, tuinbouw / horticulture, boeren / farmers) 9. Transportation Coded sub-categories: a. Infrastructure (infrastructuur / infrastructure) b. Road traffic (verkeer / traffic, wegen / roads) c. Schiphol / aviation (Schiphol / Schiphol Amsterdam Airport, luchtvaart / aviation, vliegveld Lelystad / Lelystad Airport) d. Public transportat (openbaar vervoer / public transportation, treinen / trains, bussen / busses) 10. Housing Coded sub-category: a. (Public) housing and housing market(wonen / living, sociale woningbouw / social housing, hypotheekrente / mortgage interest, woningmarkt / housing market, huizenmarkt / housing market) 11. Education and Culture Coded sub-categories: a. Education and science (onderwijs / education, school / school, wetenschap / science) b. Art and culture (cultuur / culture, kunst / art) C. Political leaders C1. Which political leader is/are mentioned in the item? 1. PVV: Geert Wilders 2. VVD: Dilan Yeşilgöz-Zegerius 3. NSC: Pieter Omtzigt 4. GroenLinks-PvdA: Frans Timmermans 99. None 71 D. Professionalisation D1. What level of personalisation does the item have? Note. D1 is only coded if the content type (A4) is an advertisement. 1. General (targeting a broad age audience) 2. Personalized (targeting specific age demographics) D2. What are the demographic targeting practices of the item? Note. D2 is only coded if the content type (A4) is an advertisement. 1. Specific demographics (zipcode) 2. Broad audience (countrywide) D3. Is there a call-to-action on the item? Note. The following reference words are used to see if the item meets the call-to-action criteria = stem / vote and kies / choose 1. Yes 2. No D4. What is the spending per item? Note. D4 is only coded if the content type (A4) is an advertisement. 1. €0 - €100 2. €100 - €199 3. €200 - €299 4. €300 - €399 5. €400 - €499 6. €500 - €599 7. €600 - €699 8. €700 - €799 9. €800 - €899 10. €900 - €999 72 11. €1000 - €1500 12. €1500 - €2000 13. €2000 - €2500 14. €2500 - €3000 15. €3000- €3500 16. €3500 - €4000 17. €4000 - €4500 18. €4500 - €5000 D6. What is the total number of likes, comments, and shares on Facebook posts? Note. D6 is only coded if the content type (A4) is a post. 1. 0-499 2. 500-999 3. 1000-1499 4. 1500-1999 5. 2000-2499 6. 2500-2999 7. 3000-3499 8. 3500-3999 9. 4000-4499 10. 4500-4999 11. 5000-5499 12. 5500-5999 13. 6000-6499 14. 6500-6999 15. 7000-7499 16. 7500-7999 17. 8000-8499 18. 8500-8999 19. 9000-9499 20. 9500-9999 21. More than 10.000 73