Comparing climate change news media discourses : a storyline approach to analyse the climate change discourse of two Dutch regional newspapers

A storyline approach to analyse the climate change

discourse of two Dutch regional newspapers

COMPARING CLIMATE CHANGE

NEWS MEDIA DISCOURSES

Author:

Lotte Molenaar

(11007508)

Supervisor: Jaap Rothuizen

2

_{assessor: Fred Zaal}

Bachelor Thesis

Human Geography

Acknowledgements

This thesis is the end product of both my major program in Human Geography, as well as my overarching bachelor program in Future Planet Studies, both at the University of Amsterdam. Starting this thesis project was exciting, as it felt a bit like ‘the beginning of the end’. However, in spite of this promising courage, it turned out to be quite challenging to actually reach this endpoint, because if there is one thing I learned during the process, it is what it means to do iterative research: Analysing data over and over again, adjusting and readjusting -and readjusting. I will never in my live have to Google this concept again. Nevertheless, slowly, the finish line appeared. And we made it, as you have the proof in your hands. But I could not have done this all by myself.

First of all I owe big thanks to my thesis supervisor, Jaap Rothuizen, who supported me during the entire process; who was willing to spruce with me on my thoughts, or came up with news input from his side; who was willing to give feedback on my work even though I send it in on random moments; and who, most importantly, made me see what reasonable goals were within the given timeframe.

Furthermore I would like to thank the interviewees: Annemieke van Dongen, Peter Bruijns and Arie Leen Kroon. Thank you for making time for me, and the overall enthusiasm towards me. Our conversations made my thesis come alive, and your information was very valuable with regard to the context of my research and added to my understanding of the climate storylines in relation to the regional newspapers.

Also I am very grateful towards my uncle and aunt, Marcel Molenaar and Aideen O’Rourke, who were crazy enough to agree on going over my thesis for a language check and some last feedback.

Finally, I would like to thank my friends and family, who were willing to listen to me and supported me when needed, but also laughed at me and reminded me of the fact that no more than a handful of people would ever really read this thesis; putting me with my feet back on the ground again and see everything in perspective.

Abstract

Over the last decades the concept of climate change is rising within newspaper coverage worldwide, encouraging public concern on the matter. However, regional differences appear regarding the amount of attention and the content of newspaper articles written on climate change, also within the Netherlands. This deviation is often explained by the specific discourses per newspaper with regard to the topic, which is partly determined by their context. The objective of this study is to establish a more extensive understanding on the phenomenon of climate change news media discourses, and to provide insights in the coherent role of contextual aspects, by exploring how the climate change news media discourse of two regional newspapers, the Provinciale Zeeuwse Courant (PZC) and Dagblad De Limburger (DDL), differ from each other during the first three months of 2019. To conduct this qualitative comparative case study research, all climate change incorporating articles of both regional newspapers published in the first quarter of 2019 are extensively analyzed in order to uncover the climate-storylines of both newspapers. These storylines are used, in combination with additional data from three in-depth interviews, as a tool to reduce discursive complexity, to uncover the climate change discourse for both the PZC and DDL.

By means of the (water)basin-storyline regarding the PZC, and the weather-storyline for DDL, it is argued that although both regional newspapers start from the same assumptions in relation to climate change, they develop a different discourse. The PZC emphasizes the aspect of ‘action’ in relation to issue, as to mitigate the threatening future implications of climate change for the province. DDL on the other hand focusses on ‘adaptation’ strategies, as the implications are already occurring, and will only increase. The meanings are embedded within the concept of climate change itself through the sign-network built around the concept by the newspapers in interaction with their readers, and represent the climate change news media discourses.

Table of Contents

Acknowledgements ... 1

Abstract ... 2

Table of Contents ... 3

Abbreviations ... 5

List of Figures ... 5

List of Tables ... 5

List of Charts ... 5

Introduction ... 6

Theoretical Framework ... 9

1.

The fundamentals of discourse ... 9

1.1 Language ... 9

1.2 Social Constructionism ... 10

1.3 Defining Discourse ... 11

2.

Climate Change ... 11

2.1 The Climate ... 11

2.2 Context and Meanings ... 12

3.

Representation through news media ... 13

3.1 Agenda-Setting ... 13

3.2 Regional Newspapers ... 14

3.3 Storylines ... 15

Methodology ... 17

1.

Research strategy ... 17

2.

Research design ... 17

2.1 Case: Provinciale Zeeuwse Courant (PZC) ... 18

2.2 Case: Dagblad De Limburger (DDL) ... 19

3. Operationalization ... 20

3.1 Units of analysis ... 21

4.

Data collection ... 21

5.

Data analysis ... 23

5.1 Attention ... 23

5.2 Storylines ... 24

5.3 Discourse ... 26

Results ... 27

1.

Amount of Attention ... 27

2.

Article Content ... 30

2.1 The Themes ... 31

2.2 Climate ... 32

2.3 Society ... 33

2.4 Economics ... 34

2.5 Politics ... 35

2.6 Culture ... 36

2.7 Science ... 37

Analysis ... 39

1.

Storylines ... 39

1.1 Indicating the Storylines ... 40

1.2 PZC: The (Water)Basin-Storyline ... 41

1.3 DDL: The Weather-Storyline ... 44

1.4 The Youth-Storyline ... 47

2.

Discourse ... 48

2.1 Communication through Storylines ... 50

2.2 Action versus Adaptation ... 52

2.3 Connotations ... 54

2.4 The Impact of Discourse... 56

Conclusion ... 59

1.

Findings ... 59

1.1 Proviciale Zeeuwse Courant ... 59

1.2 Dagblad De Limburger ... 60

2.

Limitations... 60

3.

Recommendations ... 61

Bibliography ... 62

Appendix ... 66

1.

Article Collection Provinciale Zeeuwse Courant ... 66

2.

Article Collection Dagblad De Limburger ... 68

Abbreviations

PZC Provinciale Zeeuwse Courant DDL Dagblad De Limburger

List of Figures

Figure 1: Coverage of Climate Change in Worldwide Sample of Newspapers (Holt &

Barkemeyer, 2012). ... 6

Figure 2: The Roots of Discourse. ... 10

Figure 3: Global Annual Temperature Anomaly (°C) (NASA, 2015). ... 12

Figure 4: Place Indication of the Regions in which the Provinciale Zeeuwse Courant and Dagblad De Limburger are published. ... 20

Figure 5: Research Questions. ... 20

Figure 6: Coherence between the Themes. ... 31

Figure 7: PZC Article Contribution to the Basin-Storyline during the First Three Months of 2019. ... 42

Figure 8: DDL Article Contribution to the Weather-Storyline during the First Three Months of 2019. ... 45

Figure 9: PZC and DDL Article Contribution to the Youth-Storyline during the First Three Months of 2019. ... 48

List of Tables

Table 2: The Climate Theme. ... 32

Table 3: The Society Theme. ... 33

Table 4: The Economic Theme. ... 35

Table 5: The Politics Theme. ... 36

Table 6: The Culture Theme. ... 37

Table 7: The Science Theme. ... 37

Table 8: Storyline Word-Count Analysis. ... 41

Table 9: Index Research Attachments...

69

List of Charts

Chart 2: Number of Articles Published per Month. ... 27

Chart 3: Number of Articles Published per Week. ... 28

Chart 4: Number of Published Articles per Day. ... 29

Introduction

We, humans, are influenced by news media (Castells, 2013; Dayrell, 2018; Entman, 1989, Sampei & Aoyagi-Usui, 2009; Smith, 2005). They do not determine exactly what we think, but do have an impact on what we think about, by discussing and perpetuating concepts, ideas and ideologies (Dayrell, 2018; Entman, 1989). Because of news media we, as ‘the public’, learn about events, proceedings, affairs, scientific findings and social implications (Castells, 2013). A positive correlation is shown between an increase in printed news media coverage on a topic or issue, and an increase in public concern for this issue (Eveland & Dunwoody, 2002; de Waal & Schoenbach, 2008; Sampei & Aoyagi-Usui, 2009), which explains how an issue can be moved from being perceived as an aspect or condition, into a public issue, and often even up to a policy concern (Castells, 2013; Rogers & Dearing, 1988).

A rising concept in global news media coverage is ‘climate change’ (Boykoff, 2007; Holt & Barkemeyer, 2012). The concept is not only recurring more often in printed news media, but is also reflected through its increasing prominence in films, documentaries, and even educational curricula (Tilbury, 1995). Focusing on the worldwide newspaper coverage of climate change, the study of Holt & Barkemeyer (2012) shows that there is an significant increase in climate change news articles since the 1990s (figure 1). The study of Sampei & Aoyagi-Usui (2009) also confirms that the dramatic increase in climate change newspaper coverage in Japan resulted in an elevated public concern on the issue of global warming.

As climate change is likely to affect the planet and our livelihood (IPCC, 2007; IPCC, 2014), public understanding and concern on this topic is important to accomplish changes which are beneficial with respect to slowing down the process of global warming (Castells, 2009). However, news media are selective in what they publish, resulting in the appearance of regional differences regarding the coverage of climate change in news articles (Vasterman, 2004). Newspapers in Australia and the United Kingdom for example published many more climate change news articles compared to newspapers in the United States of America (Holt & Barkemeyer, 2012). The context in which news media operate, incorporating aspects such as politics, culture, nature, demographics, economy, geography, history etc., is considered important in relation to these variations (Boykoff, 2007; Hajer & Versteegh, 2005; Jørgensen & Phillips, 2002).

Regional differences might be distinguishable on a national level as well: When the concept of climate change is analysed within various Dutch regional newspapers for example, the news media coverage on this topic is shown to be very diverse (LexisNexis, 2019). However, instead of only analysing the differences in attention by the number of articles on the matter, the content could be analysed as well. The content of an article, and most definitely the content of a sequence of articles which address the same topic, are likely to refer to the context and underlying assumptions of the newspaper, through which can be explained why differences in news media coverage are existing (Jørgensen & Phillips, 2002). To analyse the content of multiple newspaper articles on climate change, studying their discourses regarding the topic is argued to be a suitable method (Carvalho & Burgess 2005; Dayrell 2018). Discourses can be explained as the interrelation between texts through which the audience can assign meaning to a socially constructed concept (Hajer, 1995; Jørgensen & Phillips, 2002). By doing a discourse analysis, a connection between the climate change news articles published by one regional newspaper can be understood, through the interrelation between the contents of these articles.

Because news media influence their audience, newspaper discourses on climate change will influence the way their readers perceive climate change, by means of the meanings they attribute to this concept effectuated through discourse (Carvalho & Burgess, 2005; Dayrell, 2018; Hajer, 1995). Therefore, understanding the climate change news media discourses of regional newspapers, issued within another province of The Netherlands can provide valuable insights in the role of contextual aspects in relation to news media, and how readers of the newspapers within these regions might perceive climate change differently, as they are influenced by a different medium.

The two regional newspapers that will be studied in this thesis are the ‘Provinciale Zeeuwse Courant’ (PZC), issued in the province of Zeeland, and ‘Dagblad De Limburger’ (DDL), issued in the province of Limburg. By examining these newspapers, this study will explore:

“How the climate change news media discourses from the Provinciale Zeeuwse Courant and Dagblad De Limburger differ from each other during the first quarter of 2019”. The

theoretical framework will elaborate on the relevant concepts in relation to this research topic, after which the methodology chapter will explain how data is collected in line with the research strategy, and clarify how the relevant concepts will be operationalized in order to analyse the data. The results and execution of this data analysis are described in the successive chapters, until the study will finalize with the conclusions and recommendations, which reflect upon the most notable findings with regard to the theoretical background or societal impact, and possible opportunities for future research.

Theoretical Framework

In order to be able to analyse climate change discourses in different news media, there must be an extended understanding on the concept of discourse in itself (Hajer, 1995). In line with this statement, discourse will be the focal point within this entire research: It will serve as the theory and a large part of the method, which explains why this theoretical framework is built around discourse theories.

1. The fundamentals of discourse

Although discourse is a widely used concept, and discourse analyses are conducted in a wide range of research, the concept is fairly complex and is underpinned by a large underlying structure rooted in linguistics (Jørgensen & Phillips, 2002). Besides, there are different perceptions on how it precisely is, should, or could be used, which again is also dependent on the discipline and context. This means that not a single unambiguous discourse theory exists (ibid.). However, mainly when going back to the basic principles of discourse, there are certain commonalities among the discourse theories of important philosophers and scientists. Yet, before going into these fundamentals or ‘core’ of discourse theory, it is useful to state at least one, very broad, definition of discourse, which will give an idea of the direction this premise is moving towards. Jørgensen & Phillips shortly define discourse as: “A particular way of talking about and understanding the world (or an aspect of the world).” (2002 p.1). Keeping this definition in mind, a discourse

analysis can be seen as a tool to uncover social structures or patterns with regard to a certain topic or debate in society.

1.1 Language

Now that it is clear where about to end, we can start at the beginning: language. Language is our human way of construction, as we build and adjust our world by the use of language (Gee, 2004). Technically, we cannot simply understand objects around us, we can only grasp them by ascribing a certain meaning to them. So, when we speak or write about the world around us, we create conceptions to give meaning to what we see. By this we use language to construct reality, and, in this way, make reality accessible to ourselves (Jørgensen & Phillips, 2002). This basis of discourse theory originates from the structuralist and post-structuralist ideas in linguistics philosophy. ‘Linguistics’ speaks of signs: each word is a sign, and signs get their meaning through the entire sign-network, thus not through an actual connection with the real world. Culler (1986) explains in his book, dedicated to structuralist linguistic scientist Ferdinand de Saussure, the difference between ‘la langue’ and ‘parole’, where ‘la langue’ is explained as the above mentioned sign-network, which is seen as the fixed structure of language, and where ‘parole’ is seen as the human’s concrete use of signs, by which it becomes subject to change. Although

post-structuralism agrees on the structuralist theory that the meaning of signs does not come from reality, but lies within the sign-network, it differs from structuralism by saying that language is changeable: Meanings and structures are temporary and can change, for example because of context or through time (Jørgensen & Phillips, 2002). ‘La langue’ and ‘parole’ are not two separate categories anymore, but they overlap, leading up to the post-structuralist vision that language is always ambiguous: It is structured by patterns and discourses, which can be altered through ‘discursive practices’, which are explained by Foucault as actively produced social-psychological realities (Foucault, in Hajer, 1995). These central assumptions of post-structuralism underlie discourse theories (Jørgensen & Phillips, 2002).

1.2 Social Constructionism

So, assigning meaning to anything through language is what makes us understand, or, some even say, create reality. When zooming out a little, it can be said that we use meanings to structure, categorize, shape and change our world, identity, and social relations (Jørgensen & Phillips, 2002). This perspective is known as social constructionism, a society and culture theory which can be seen as a broader category, in which post-structuralism also belongs. Social constructionism entails that we as humans (i.e. the ‘social’) actively construct the world (or ‘reality’), based on our identity, knowledge, outlook, culture, history and so on. This results in a world which is ever changing, and does not possess any objective truths, as physical ‘truths’ are depending on interpretation, translating them it into the subjective spectrum (Burr, 1995). However, in our everyday world we do shape and perceive truths, based on knowledge exchange and social acting. Some ideas become normal or logical, while others are positioned as not, by which social constructionism is placing restrictions on the constructed world (Jørgensen & Phillips, 2002). This is where discourse theory finds its roots, or where the loop can be closed: 1) We create, structure, and alter our world through language. 2)

Language is understood because a meaning is assigned to everything. 3) Any meaning is gained through discourse. 4) Discourse is created through common views or truths, based on the communication of knowledge and ideas. 5) This communication or exchange is embodied through language (figure 2). This background information on the fundamentals of discourse makes that the simple discourse -definition by Jørgensen & Phillips gets more substance.

1

2

3

4

5

1.3 Defining Discourse

As a consequence of the knowledge accumulation on the topic of discourse, it is conducive to reason from a more sophisticated and detailed definition of discourse in this research. Used definitions by scientists or in studies will always slightly deviate from each other, but by now it is elucidated that their origin is the same. Hajer & Versteegh define discourse as: “An ensemble of ideas, concepts and categories through which meaning is given to

social and physical phenomena, and which is produced and reproduced through an identifiable set of practices.” (2005 p.175). The linguistic roots shimmer through in this

description, just as the social constructionist ontology. This definition does not mention ‘the world’ or ‘reality’, but refers to ‘phenomena’, with the assumption that the world is no more than a system of (socially constructed) phenomena. This is relevant because climate change is a perfect example of a socially constructed phenomenon; what means that for its essence and entire existence, climate change is dependent on discourse (Hajer, 1995).

2. Climate Change

Climate change as a concept is emergent in a substantial amount of scientific literature, debates, (news) media and television shows. However, the concept does not always imply entirely the same thing, as it is presented within a certain discourse (Hajer & Versteegh, 2005). This supports the linguistics’ assumption that it is impossible to objectively assign meaning to a phenomenon. Although it is important to keep this in mind, a certain climate change ‘foundation’ must be established to be able to build further on this concept. The definition of the Fifth Assessment Report from the Intergovernmental Panel on Climate Change (IPCC) will therefore be used: “Climate change refers to a change in the state of

the climate that can be identified (e.g., by using statistical tests) by changes in the mean and/or the variability of its properties, and that persists for an extended period, typically decades or longer. Climate change may be due to natural internal processes or external forcings such as modulations of the solar cycles, volcanic eruptions and persistent anthropogenic changes in the composition of the atmosphere or in land use.” (IPCC, 2014

p.120).

2.1 The Climate

For the first sentence of this definition it is important to understand that climate is explained as the average weather (IPCC, 2014). As the weather is constantly changing, actual changes in the state of the climate might be perceived as slow developments. However, climate is subject to change, just as it always has been. This is illustrated by the last million years in which the world’s state of climate is characterized by a sequence of ice ages, with warming periods in between (Berg, 2007). After the last little ice age in the 18th century, we currently find ourselves in a warming period, indicated by the rising global mean temperature over the last 130 years (figure 3). The measurements of four

different institutes show matching results and indicate a global temperature rise of 0.8 degrees Celsius between 1880 and 2015, of which 65 percent occurred since 1975 (NASA, 2015).

The second sentence of the IPCC definition says that climate change may be due to natural variability, or is induced by human activities. So, although climate is changing as long as the earth exists, humans might have a stake in the current process of global warming as well: Since the industrial revolution in the second half of the 19th century, we started to emit large amounts of greenhouse gases, e.g. by burning fossil fuels, into the atmosphere (Berg, 2007). Despite the fact that we need greenhouse gases in the atmosphere to create a liveable earth-temperature, an amplified greenhouse effect seems to be the result of the extra emissions (ibid.). Sceptics and alternative thinkers question this anthropogenic point of view (Urry, 2015), but the 2007 Climate Change IPCC report argues that there is a 90 percent confidence that the ongoing global warming is caused by human activities. This means that a certain consensus on this matter is established within the scientific community (IPCC, 2007).

2.2 Context and Meanings

However, the whole (anthropogenic) climate change issue is not considered important because of its concrete existence or not. The physical phenomenon of climate change only gained relevance through the way society designates meaning to it, as the social construct is what we understand (Hajer & Versteegh, 2005). The reason why, at some point, we start to make sense of a phenomenon, is related to time, place and attention (Hajer & Versteegh, 2005). Hajer (2002) explains this by the use of dead trees: Dead trees in itself are a physical phenomenon, not a social construct, but when people come across dead trees during a time of acid rain discussions, a specific meaning is given to these trees. This happens because societal experiences shape how we think and act, which in this case has the effect that dead trees become a social phenomenon, and through this they become a metaphor, signifying to the larger problem of acid rain and its causes. So, at present,

climate change has developed into a social phenomenon, but what exact meaning is given to the concept depends on the context.

Following Hajer’s (2002) train of thought, the developments in attention for climate change at the international level could for example be explained by the disclosure of the increasing rise in global annual temperature and its expected implications on environment and society (NASA, 2015). Through this structure of presentation, the physical phenomenon of climate change received a negative connotation. The concept of connotation is explained as the feeling or idea that a concept invokes in addition to its denotative (or literal) meaning, and is often shaped through collocates which carry a certain value (Feng et al., 2013; Sinclair, in Orpin, 2005). Furthermore, as the increased greenhouse gas emissions are seen as the main culprit of the phenomenon (Berg, 2007), climate change now signifies to a human induced issue. This is a simple representation of how climate change could be transformed into a social phenomenon with coherent meanings. However, the way climate change is exactly perceived and what meanings are assigned to it, differs from time to time and place to place, as with the assumption that our reality is a system of socially constructed phenomena, the context is always determinative (Hajer & Versteegh, 2005). Referring back to the idea that we create certain ‘truths’ on which discourses rely, these truths are based on historical, political, cultural and geographical context (Jørgensen & Phillips, 2002; Hajer & Versteegh, 2005). This explains that climate change discourses can even deviate from each other within one country, for instance between regions: The implications of climate change in a coastal area (e.g. the threat of sea level rise) (Nicholls & Cazenave, 2010), are for example very different from the implications for regions who experience dry seasons (e.g. the threat of desertification) (Millán et al., 2005). These different contextual aspects make that the two regions have a different ‘story’ or perception on what is important or relevant, by which the metaphor of climate change will signify different meanings, through which divergent truths are created.

3. Representation through news media

Hajer (2002) demonstrated with the dead trees example that we only make sense of a phenomenon because it receives certain attention, causing it to ‘suddenly’ cross our path and becomes an experience which is worthwhile or necessary to assign meanings to. News media play an important role when it comes to attention, as it is an agenda-setting practise, on multiple levels (Rausch, 2002).

3.1 Agenda-Setting

The first level is represented by the setting of the media agenda itself, where topics come through when they transcend the so-called distinct media gatekeepers’ criteria. This agenda is mainly formed by newsworthy and spectacular events, press releases and results

of various research reports. Subsequently, the spread of this media agenda makes that the public agenda is created through the subtle process of issue communication, mind persuasion and idea formulation, which embodies the second level of agenda-setting. Thirdly, there is the policy agenda, which is generated through the synergy between the general public and policy makers. This policy agenda can also directly be affected by the media agenda, and in return the policy agenda can influence the media agenda (Rogers & Dearing, 1988; Vasterman, 2004; Kepplinger & Habermeier, 1995). These agenda setting elements are entangled within the three news media serving functions distinguished by Vasterman (2004): the news-report function; the platform function; and the ‘guard-dog’ function. The first function refers to the value of being informed on issues, events and information. Secondly, news media function as a hatching mechanism that spread the ideas of different actors on different topics. Finally, news media functions as a guard-dog, because they closely follow the government, organisations and institutions. However, issues, events, and information are in an attention-competition, as news media cannot discuss everything at the same time, and new stories crowd out the older ones (Rogers & Dearing, 1988). This means that news media handle a selective provision of knowledge, which is specific to each news medium in itself (Carvalho & Burgess, 2005; Vasterman, 2005).

3.2 Regional Newspapers

According to the Chief Editor of the regional newspaper in the province of Zeeland, regional news media report on: “All news that people living within the region have to know about.” (Kroon, pers.com. 2019). This means that the publication on news events is not determined by the borders of the applicable region (an event does not have to happen within the region itself to be relevant news), but is depending on the implications or consequences of the particular news event in relation to the people located in the region. So, the attention-competition for regional news media is partly dependent on the relevance of a topic regarding the specific region. The borders of a region are generally defined by the news medium itself, as they can vary from for example national television broadcasts or newspapers, to radio broadcasts which are relevant to only one or a few townships. The distribution area of a regional newspaper can for example be determined by the borders of a province, as this is within The Netherlands inter alia applicable to the province of Gelderland, Brabant, Noord-Holland, Limburg and Zeeland (NOM, in Mediamonitor, 2017). A regional news medium does not intend to address only a single target group in whatsoever kind of category (e.g. regarding political preference or level of education), because they mean to serve all the people living within their region: They have to consider a very broad audience (Bruijns, pers.com. 2019; van Dongen, pers.com. 2019; Kroon,

pers.com. 2019). However, printed regional newspapers in The Netherlands are currently

category 50 years and older (van Dongen, pers.com. 2019). Another distinctive aspect of regional newspapers is that they want to really profile themselves as regional newspapers, which is represented by the various pages per newspaper which solely publish news that comes from within, or contains a link with, the region (Bruijns, pers.com. 2019). But the Dutch regional newspapers also collaborate as they have the possibility to exchange articles among each other, which is needed because they themselves do not generally have news reporters all over the country (van Dongen, pers.com. 2019). The choice of news items to be published on a topic, or how they are framed, depends on the preferred ideological worldview of the particular news medium (Carvalho & Burgess, 2005; Vasterman, 2004). This worldview is effectuated by ‘storylines’, implying that if the medium wants to publish an article, in this case related to the topic of climate change, the scope of the article must somehow fit in the existing storyline(s) of this medium regarding the topic. Because of this, the established assumptions and ‘truths’ on the matter work through in the accumulation of articles, ensuring that each report adds something to climate change story of the particular news medium (Hajer & Versteegh, 2005; Rogers & Dearing, 1988).

3.3 Storylines

News media never speak directly about issues; they talk about stories, as stories are what interests people (Radford, in Smith, 2005). Transforming complex issues into narratives is how information is represented and communicated through media. With the formulation of such stories news media actively participate in the social construction of reality, which explains their influence on forming, erasing, and communicating discourses (Allan et al., 2000; Smith, 2005). When more stories on the same topic are published within a news medium, a storyline is created, to which newly published articles will contribute. This storyline concept is drawing attention because it can be used as a tool to understand a complex interplay of discourses (Smith & Kern, 2009; Davies & Harré, 1990). Hajer describes storylines as: “Narratives on social reality through which elements from many

different domains are combined and that provide with a set of symbolic references that suggest a common understanding.” (1995 p.62). A recognizable example of how this

works in practise can be obtained from news items in relation to refugees: Whenever refugees are discussed, migration issues will be mentioned or shimmer through within the item. The underlying migration-storyline is uncovered in this way, and allows people (unconsciously) to place the (new) news item within the existing (political) context of the theme.

The power of a storyline approach lies within the metaphor effect, as through the story a linguistic element which could receive any discursive meaning, is caught within a context, and therefore signifies to the storyline as a whole (Hajer, 1995). This phenomenon was

exemplified by the previously discussed dead trees which became a metaphor for the overall story on acid rain (Hajer, 2002). By means of the metaphor effect, discursive complexity is reduced, and any actor can adjust to it in its own way and position themselves with respect to the discussed issue (Hajer, 1995). With regard to news media usage of storylines, the precondition to make it credible is that it should ‘sound right’, as it mimics the format of daily communication practices. Also, the storylines should be relatively easy to pick up on. Making something sound right does not only depend on the arguments, but also relies on the additional feelings (embedded within the connotations) and on who is formulating the storyline, as trust towards the news medium is a key factor for the acceptance (ibid.). Because storylines can be regarded as a part of the construction of social phenomena, the way news media develop them is determined by the contextual aspects in which a medium operates (Jørgensen & Phillips, 2002). A storyline is not static: changes or transitions can evolve, and more than one storyline can be related to one topic. Storylines make that the news media recipients work out a context in which they can place and understand an issue: It allows them to assign meanings. A storyline comes forward through the underlying assumptions and truths regarding the news report, and discourse is taken up within the arguments, language and assumptions of the storyline(s) (Hajer, 1995). So, storylines can play an essential role in understanding discourses within news media, as they form the context and handles for any new news item related to the topic.

Methodology

1. Research strategy

When researching regional climate change news media discourses, the assumptions related to the social constructionist ontology of discourse are directive when it comes to the decision of research strategy (Bryman, 2012a). This constructionist approach implies that reality only exists because of our human perception on phenomenon, which means that nothing is ever static, and everything is dependent on our interpretation. To be able to understand any process or phenomenon within this socially constructed world, a qualitative research strategy is required, as qualitative research focuses on ‘why’ and ‘how’ certain phenomena, characteristics or relations exist. It is an explorative approach to obtain more in-depth knowledge and a greater understanding of the research subject (ibid.). The subject of this thesis is regional climate change media discourses, which means that the emphasis is on discourse-analysis. A complex characteristic of discourse-analysis is that theory and method are somewhat intertwined (Jørgensen & Phillips, 2002). So, although qualitative research is normally conducted in an inductive manner, meaning that the researcher works towards formulation of theory, this study starts from theory as well, as formulated in the theoretical framework. However, induction will still be present, as the data obtained from the units of analysis will immediately be theorized by interpretative descriptions, which will be written down in analytic memos. When all units of analysis are analysed in this way, the data and descriptions will be analysed again, to verify if new theory can derived (based on for example the acquired knowledge from the first ‘round of analysis’) (Bryman, 2012b). This aspect of repetition illustrates the iterative aspect of inductive research, and will continue until no more new theory is derived from the data, and a point of theoretical saturation on the matter is set (Bryman, 2012c). This analytic approach could be labelled as ‘semi-grounded-theory’: admitting that it started from theory, but still works in an iterative way and aims to be mainly descriptive and explanatory towards newly formed theory in relation to the researched phenomena.

2. Research design

This study makes use of a ‘comparative design’, what in qualitative research means that a multiple-case study is implemented, wherein the cases (eventually) will be compared to one another (Bryman, 2012d). This is a relevant research design because news media discourses of different regions in The Netherlands need to be compared to signify how they vary from each other during the first three months of 2019. In this research the cases are represented by two regional newspapers, both tied to another region. Newspapers form a suitable medium for discourse-analysis by their use of a standard format (Holt & Barkemeyer, 2012), and because the study of language can be executed in its pure shape (Pajunen, 2008). Conversations or presented texts, probably supported by gestures,

attitude and perhaps even images or video, contain a lot of extra elements which influence the propagation of discourses, while a written article has to stand by itself (Hulshuis, 2010). Furthermore, newspapers are still a frequently used news bulletin, as 47 percent of the Dutch populations reads newspapers on a daily basis (Witte & van Strien, 2018). Although the annual circulation of Dutch daily newspapers is steadily decreasing since 2008, regional newspapers, especially the ones distributed outside of the Randstad, are still published in significant numbers, by which they reach a large audience (NOM, in Mediamonitor, 2017). Taking into account that location and situational aspects influence a storyline, the two newspapers that will be analysed are published in two outer corners of The Netherlands, i.e. in the province of Zeeland and the province of Limburg. These regions are characterized by very different contextual factors, not only in comparison to each other, as well as to the whole country (CBS, 2017a-b; GeoDesk WUR, n.d.; KNMI, 2018; Geologie van Nederland, n.d.a-b). Besides, these newspapers, called ‘Provinciale Zeeuwse Courant’ and ‘Dagblad De Limburger’, represent the two Dutch regional newspapers with proportionally the largest amount of readers (NOM, in Mediamonitor, 2017).

2.1 Case: Provinciale Zeeuwse Courant (PZC)

The province of Zeeland (marked in orange, figure 4) is located in the southwest corner of the Netherlands, adjacent to Belgium and the North Sea, and counts about 382.000 inhabitants (CBS, 2017a). The province is a coastal area which is almost entirely located below NAP, and part of the Rijn-Maas-Schelde-Delta (GeoDesk WUR, n.d.). Zeeland is a marine clay landscape (Geologie van Nederland, n.d.a) and characterized by its peninsulas, disastrous flood of 1953, modern day innovative flood-defence systems, and the largest number of tourists relative to the number of inhabitants, per year (CBS, 2017b). The Provinciale Zeeuwse Courant is their daily regional newspaper published by Wegener Niewsmedia B.V. in collaboration with ADR Nieuwsmedia, which is the largest news brand within The Netherlands. They fall under the overarching media publishing organization De Persgroep, whereof the editorial office is located in Rotterdam (van Dongen, pers.com. 2019). ADR Nieuwsmedia fabricates complete news pages for several regional newspapers among which the PZC. Apart from this supply, the PZC publishes about 8 pages of solely Zeeuws news on a daily basis; is the front page hybrid; and is the PZC’s editorial team allowed to make alterations in the supplied news pages (Kroon,

pers.com. 2019). Through this mechanism the PZC effectuates that is serves as a complete

regional newspaper, but also as a sufficient alternative for a national newspaper (ibid.), as it has a regional focus, but does report on national and international news items, sports, arts, culture and human interests as well (LexisNexis, 2019). The PCZ has a circulation of 41.109 printed newspapers per issue, with which they reach an audience of 117.000 people above thirteen years of age (NOM, in Mediamonitor, 2017). Because Zeeland counts

324.000 people above the age of thirteen (CBS, 2017a), it can be calculated that about 36 percent of the people living in Zeeland reads the PZC, and they do not really experience competition from another newspaper (Kroon, pers.com. 2019). The Chief Editor describes the audience as statistically not very distinctive from the (generally middle-aged) national printed newspaper audience: they do only deviate a little regarding their interest for employment opportunities, and for religion or spirituality. However, when it comes to the nature of the people, they are a bit more reserved as a remnant from their island-history (ibid.). The profile of the PZC is broad, as they are writing for all the inhabitants of the province. No niche is chosen, they will not campaign, and one-sided writing is not an option (van Dongen, pers.com. 2019; Kroon, pers.com. 2019).

2.2 Case: Dagblad De Limburger (DDL)

The province of Limburg (marked in blue, figure 4) is located in the southeast corner of the country and adjacent to both Belgium and Germany. The province is home to 870.000 residents and did only officially become part of The Netherlands by 1848 (CBS, 2017a; GeoDesk WUR, n.d.). It is characterized by its elongated shape and stands out because of its high location: up to about 70 meters above NAP in its hilly southern part (GeoDesk WUR, n.d.). The river Maas flows through the province, and historically a lot of peat, coal, and lignite mining has taken place in the area (Geologie van Nederland, n.d.b). Currently Limburg is the most popular province for domestic tourism (CBS, 2017b). Another remarkable aspect is Limburg’s state of climate: They experience slightly more influence of the continental climate instead of the temperate maritime climate found in most parts of The Netherlands, because of their location relatively far from the sea. This result is that the Dutch’ extreme temperatures, as well in the warm as in the cold spectrum, are often measured in Limburg (KNMI, 2018). Dagblad De Limburger is published by Media Groep Limburg (LexisNexis, 2019), and is an independent regional newspaper with a strong and serious editorial team, but they do have the opportunity to exchange articles through partnerships with other newspapers (van Dongen, pers.com. 2019). DDL really profiles itself as thé regional newspaper for people in Limburg, illustrated by the first ten pages which only discuss news items which are somehow related to Limburg (Bruijns, pers.com. 2019). Afterwards other topics such as national news, foreign news and economics are published, which shows that the DDL is mainly focused on regional news (ibid.). DDL is the regional newspaper with the largest circulation number in The Netherlands: 121.000 printed newspapers on a daily basis (NOM, in Mediamonitor, 2017). 739.000 of the people in Limburg are older than thirteen (CBS, 2017a), and the reach of DDL is 352.000 people, which means that almost 48 percent of the inhabitants of Limburg reads the newspaper (Mediahuis Limburg, n.d.). They write with a very broad audience in mind: average middle-aged people with general interests. Their emphasize is on informing their readers in

relation to the news, and they aim to make the different sounds with regard to a topic to be heard, as to please the entire broad audience (Bruijns, pers.com. 2019).

3. Operationalization

To be able to find answers in relation to the main research question of this thesis, stated as: “How do the climate change news media discourses from the Provinciale Zeeuwse Courant and Dagblad De Limburger differ from each other during the first quarter of 2019?”, three sub-questions are drawn up. These sub-questions realize a step-by-step approach towards answering the main research question, making the implementation of the investigation more manageable. The three sub-questions are formulated in figure 5, which shows a visualization of the research approach:

Main

• How do the climate change news media discourses from the Provinciale Zeeuwse Courant and Dagblad De Limburger differ from eacht other during the first quarter of 2019?

Sub 1

• How does the attention for climate change issues vary among the two regional newspapers within the investigation period?

Sub 2

Sub 3

• How do the storylines underline the climate change news media discourses of the Provinciale Zeeuwse Courant and Dagblad De Limburger?

Figure 4: Place Indication of the Regions in which the Provinciale Zeeuwse Courant and Dagblad De Limburger are published.

3.1 Units of analysis

To be able to answer the research questions, elaboration on the applicable units of analysis is required (Bryman, 2012c). As the focus is on the climate change discourses of two regional newspapers within the first three months of 2019, the units of analysis are represented by all published articles within the PZC and DDL in this period, with the proviso that they meet the selected article ‘search requirements’. What these requirements are will be explained in the section on data collection. More units of analysis are represented by three in-depth interviews, conducted with two authors of the selected news articles, complemented by an interview with the PZC Chief Editor (table 1). By means of the step-by-step system among the sub-questions, the output(s) of a sub-question can function as a unit of analysis as well. This is mainly reflected within the formulated storylines, which will themselves again be analysed as to lay hands on the news media discourses of both regional newspapers.

4. Data collection

To select relevant articles with climate change coverage in the period from the 1st of January up to and including the 31st of March, 2019, the online database LexisNexis is used (LexisNexis, 2019). LexisNexis gives the possibility to enter features, and to search for articles by the use of keywords or a combination of keywords. To collect a selection of relevant articles which fit the scope of this thesis, the following keywords are used per regional newspaper in LexisNexis: climate change (klimaatverandering), climate (klimaat), climate policy (klimaatbeleid), climate agreement (klimaatakkoord), emissions (uitstoot), environment (milieu), greenhouse effect (broeikaseffect), warming (opwarming), sustainable (duurzaam), region (regio), drought (droogte), and storm (storm). The settings are modified in such a way that climate change (klimaatverandering) must be mentioned in an article. Apart from this keyword, at least one of the other keywords must be present in an article. The formula is LexisNexis is as follows:

((klimaatverandering) and ((klimaat) or (klimaatbeleid) or (klimaatakkoord) or (uitstoot) or (milieu) or (broeikaseffect) or (opwarming) or (duurzaam) or (regio)

or (storm) or (droogte))) and Date(geq(01/01/2019) and leq(31/03/2019))

These research requirements resulted in a selection of 46 articles for the PZC, and 32 articles for DDL. These amounts are reduced by the following criteria:

 All articles comprising less than 200 words are removed, because they do mainly entail promotions, invitations and requests, through which they do not contribute to the aspect of issue communication (Rausch, 2002; Rogers & Dearing, 1988).  All articles that show an over 95 percent similarity rate with another article are not

 A selected article that showed to be the cultural-calendar is left out of the dataset, as it does not represent issue communication in relation to the public agenda (Rausch, 2002; Rogers & Dearing, 1988).

 Responses or letters which are send to the newspaper by their readers are left out of the article collections, because they are not written by the regional newspapers themselves.

What remains are 36 climate change issue discussing articles for the PZC, and 22 articles published by DDL. These articles represent the main units of analysis within this research. In the appendix an overview of these articles and their primary characteristics is displayed.1_{Among these articles are still some columns, interviews and one news-special.} These are left within the data collection because, although they contain opinions or give an extraordinary amount of attention to one theme, the newspapers decided to publish it, indicating that they deem their audience to be interested on the matter. Also, these items are often in imitation of present news-topics, through which they contribute to the public agenda (Vasteman, 2004; Rausch, 2002; Rogers & Dearing, 1988).

The additional collected data is represented by three in-depth interviews with ‘key persons’ with regard to the two regional newspapers. For the interviews an item list is used, because although it is important that the respondents are free to speak and should have room to elaborate on what they think is important, it is also useful to have a certain structuring tool, to make sure the conversation will not go entirely off topic (Bryman, 2012e). This makes the conducted interviews semi-structured. The details on the respondents and the interviews can be found in table 1. The transcripts of the interviews will be consulted to achieve a greater understanding of the contextual elements regarding both regional newspapers, and the general train of thought in relation to climate change incorporating newspaper articles.2

Table 1: Details on the Interviewees.

Respondent Description Date Location Duration

Annemieke van Dongen

News reporter on inter alia climate issues with the AD (Algemeen Dagblad) and several regional newspapers: Author of several articles published in both regional newspapers during the research period.

05-05

-2019 Rotterdam 47 min

1_{For the entire articles see file 1 and 2 in the additional document, named “Research Attachments” https://drive.} google.com/drive/folders/1UO4UKWjKMKPH7At6kakWE2mYWrCetD4B.

2_{For the complete transcripts and the item list in relation to the conducted interviews, see file 5, 6, 7 and 8 in:}

Peter Bruijns

Journalist with DDL and specialist with regard to economic topics and the energy transition. Author of four of the articles within the DDL data collection.

14-05

-2019 Utrecht 39 min

Arie Leen Kroon

The Chief Director of the PZC, and also the author of three comments which fall within the PZC article collection.

15-05

-2019 Vlissingen 60 min

5. Data analysis

To be able to analyse the data in order to discuss the applicable (dis)similarity of discourse as mentioned in the main research question, an understanding of climate change news media discourses related to both regional newspapers needs to be established. The operationalization of the main concepts within the sub-questions will contribute to this understanding, as through this mechanism a step-by-step system of data-analysis is realized. An important note in relation to the data analysis is that the focus will only be on the textual content of the selected articles. If articles are supported by images, these images will be left out of the analysis, because they can send out an ambivalent message: Often they are only added to the article later on, which means that this strong tool of metaphor communication does not align with the original intention of the text (Difranesco & Young, 2011; Fahmy, 2005).

5.1 Attention

The articles are selected because they incorporate the concept of climate change, indicating that these items attribute attention to the news topic. The concept of attention refers back to the theories about the role and function of news media, in which agenda-setting is very important (Rogers & Dearing, 1988). Agenda-agenda-setting is partly determined by the emphasis and frequency of attention (Rausch, 2002), whereof the latter is analysed by a quantitative approach through the means of counting. The results are set out in charts relative to the time, to provide an overview of the variation in amount of attention between the newspapers.

However, this solely quantitative analysis does only say something about the number of articles, nothing about what is emphasized by the content. To get a first idea of the emphasis, a first read-through of the articles is performed, during which is noted if the word ‘climate-‘ is already mentioned within the title and/or introduction of the article, or not, and what (other) news-themes are addressed within the articles. These notes do immediately form the start of the analytic memos.3 _{The results of this thematic analysis}

3_{For entire analytic memos, see file 3 and 4, in: “Research Attachments” https://drive.google.com/drive}

are set out in a chart and several tables, accomplished by textual examples from the articles, which give an idea of the researcher’s way of interpretation in relation to the themes.

5.2 Storylines

Because of the large overlap between theory and method with regard to discourse analysis (Jørgensen & Phillips, 2002), the operationalization of the storyline concept is in its essence already explicated within the theoretical framework. It was explained that climate change was already discussed in news media decades before 2019. This means that the two newspapers already formed one or multiple storylines in relation to this topic, functioning as the foundation to publish the contemporary articles. What the storylines entailed at its origins will be hard to determine, as they most probably have evolved over the years. Storylines are explained as the social-reality forming narratives, which combine different domains and metaphors usage for public understanding of an issue (Hajer, 1995). A story is characterized by a beginning and an ending (Hajer, 1995; Smith, 2005). The beginning can be explained by the fundamentals on which the new news articles can be build. These are presented by the assumptions and truths that the newspaper takes for granted in relation to the topic. The end of the story is relatively uncertain, as the storyline has not ended yet. This is rather typical for the concept of climate change, as it is unknown what the actual implications will be (IPCC, 2014). These elements will not literally be discussed within the articles, but they are still present in a certain sense: as the context within a newspaper in relation to the topic, or as the underlying storyline in which the climate change discourses are entangled. Therefore, the grasp of a storyline can be helpful in understanding the complex interplay of climate change news media discourses (Davies & Harré, 1990; Smith, 2005; Smith & Kern, 2009).

Uncovering the storylines starts from the previous thematic analysis, but requires a more thorough analysis of the articles. They have been read several times more to understand how both newspaper depict the ‘core’ or causes in relation to climate change (which function as the underlying assumptions), and to recognize the connection between several articles. This can be through words, topics, significations or the discussion of implications, because those connections reveal to the presence of a storyline to which those articles all contribute. However, the segments which incorporate such information should not be pulled out of their context, which is why they are not noted separately, nor are they coded in a particular way. Thoughts and explanations on these aspects are again written down in the analytic memos, also giving the opportunity to detect the segments back within the original articles. This way of keeping the text sections within the context of the article is what makes the method of analysis complex, but does also keep it close and true to the perceptions of the real world (Bryman, 2012d). The lines of argumentation in relation to

these segments are also explored, just as the idea that certain word(group)s could signify another meaning within the context. Also, attention is given to the repetition of certain concepts and the use of symbols or possible notable metaphors, as all these elements can help to uncover the fundamentals of the storylines (Hajer & Versteegh, 2005; Hajer, 1995). The other empirical method of data collection (the in-depth interviews), does also contribute to detecting the storylines, as the acquired information gives insight into the context and audience of both regional newspapers, making it easier to understand where their emphasize is in relation to the topic of climate change. References to aspects of, or information from, the interviews are made on various places within the analysis, functioning by their conjugation as a substantiation regarding the uncovered storylines (Bryman, 2012e).

The perception of these elements has evolved during the process of analyzation, as during the second read-through different aspects surface in the articles than during following thorough read-throughs, because the accumulation of acquired knowledge and perceptions influence the way the data is perceived. Overall must of be acknowledged that the output of this way of text analyzation is very person-dependent, explained by the interpretative nature of this qualitative research method (Bryman, 2012b). Although the intention is to be as objective as possible and, as a researcher, keep the personal thoughts, ideas and values aside, it is impossible to entirely maintain this aim: This is actually explained by the concept of discourse in itself, as a person can only understand and assign meaning through his or her own established sign-network, based on background knowledge, roots, environment and so on (Jørgensen & Phillips, 2002). Particular personal values and the way that the data is perceived exactly can therefore not be seen separately from the researcher, as what is apparent to one person, might not be considered that important or striking by somebody else. Besides, the analysis is based on certain choices, again made by the researcher. One of these choices is that the focus is on identifying one main storyline for each regional newspaper in relation to the climate-theme. The main storylines that will arise from the thorough and overall article analysis, contributed by the information obtained through the interviews, will be substantiated by extensive explanation and examples. Also, the researcher’s train of thought behind the choices can be recovered in the analytic memos. However, the same date could still lead to a different storyline as it is interpreted by someone else, illustrating that in spite of the transparency by the analytic memos and transcripts of the interviews, the reliability and replicability of this research is still affected (Bryman, 2012d). On the other hand, because the text of the newspaper articles cannot be altered or influenced, the reactivity is low. Also, the measurement- and ecological validity are high, as the actual socially constructed world around us will be explored within this research (ibid.).

5.3 Discourse

These interpretative aspects and their consequences are again applicable during the analysis of the actual climate change news media discourses, as the emphasis within this level of analysis is even more on meanings and feelings which do not necessarily come from the literal text fragments (making them strongly dependent on interpretation). With the identified storylines and their attributed meanings and implications in mind, all the articles will be analysed again, with a special focus on the articles which do not primarily contribute to the storylines, but who are influenced by, or relate to, the meanings that shimmer through by means of the storylines: The impact of discourse is indicated through these practices. By this the storylines are really used as a tool to reduce discursive complexity, intending to grasp the underlying climate change discourses, as those clarify the meanings with respect to the concept of climate change in the context of the regional newspapers (Hajer, 1995). The overall process of analysis effectuated by the three sub-questions assists in this formulation of climate change discourses for each regional newspaper, because it establishes the understanding of the underlying sign-network around climate change for both the PZC and DDL. The discourses can now be compared with each other to explain how they distinguish themselves, illustrating the contribution of this data analysis approach to the main aim of the research.

Results

In imitation of the newspaper article data collection requirements, the outcome is represented by 58 articles4_{. This means that during the first three months of 2019 both} regional newspapers, the Provinciale Zeeuwse Courant (PZC) and Dagblad De Limburger (DDL), together published a total of 58 newspaper articles in which climate change is discussed. The 58 articles are not distributed equally between the newspapers, as the PZC published 36 of them, and DDL 22. This article share is visualized in chart 1.

1. Amount of Attention

The variety in the amount of published articles can further be visualized per month, week and day. Starting with the article distribution numbers per month (chart 2), the results illustrate that DDL never transcends the PZC in the amount of published climate change incorporating articles. In the month January 20.6 percent of all the articles are written, wherein each regional newspaper takes an equal share. In both February and March 23 articles are published, each representing 39.7 percent of the total amount. Both months PZC published more in comparison with DDL. However, DDL shows an increase in number of climate change related publications during the three months of investigation, while the PZC shows a more fluctuating pattern in this respect.

4_{See Appendix 1 “Article Collection Provinciale Zeeuwse Courant” and Appendix 2 “Article Collection Dagblad De}

Limburger” for the summary data which is used to create the charts in this chapter.

62.1%

37.9%

Article Share per Newspaper

January February March

Number of Articles Published per Month

DDL

Chart 1: Article Share per Newspaper.

A bar graph visualization of the number of publications per week incorporating both newspapers creates a more detailed image of the diversity in the amount of attention per regional newspaper for climate change climate change (chart 3). Standing out most are the two peaks of six articles within one week, in week six and week ten, and one peak of four articles, with reference to week seven, regarding the PZC. This inter alia shows that one third of the 36 PZC articles are published in only two out of thirteen weeks of investigation. DDL does not show extreme outliers with regard to a high number of publications per week. Both regional newspapers have one week were no climate change related articles were published: week two for the PZC and week nine for DDL. Another striking observation accounts for the first couple of weeks in 2019, which largely represent the month January. Within the first week, both newspapers publish multiple articles related to climate change. However, after this week the amount of attention drops as in the following two weeks a maximum of one publication per newspaper is visible. Another ‘attention drop’ shows in week nine, running from the 25th _{of February up until the 3}rd _of March. During the last four weeks of the investigation period both newspapers publish multiple articles related to climate change again.

A daily overview displays precisely on what weekdays and dates all the articles are published (chart 4). Because the research period covers 90 days in total, the monthly graphs contain a lot of information within one visualisation. One example is an observation regarding weekdays: Both regional newspapers are issued on Mondays, but none of the climate change related articles is published in a Monday newspaper. The favourite publication day is Saturday, with twelve articles by the PZC (33.3 percent), and six articles by DDL (27.3 percent). For the remaining days (Tuesday to Friday) DDL is distributing the

3 0 1 2 3 6 4 2 1 6 2 3 3 2 1 1 2 2 1 2 2 0 3 3 2 1 0 1 2 3 4 5 6 7 1 2 3 4 5 6 7 8 9 10 11 12 13 N u m b er o f Art icl es Weeks

Number of Articles Published per Week

Provinciale Zeeuwse Courant

Dagblad De Limburger

articles evenly, with four publications on each day, while the PZC shows a fluctuating pattern with a range from two to seven articles. Chart 4 also illustrates when both newspaper publish a climate change mentioning article on the same date. This can be a ‘coincidence’, but in some cases the articles discuss the same topic, or it even concerns the exact same news article. These latter situations are interesting when it comes to the divergence in attention for a certain topic, as it shows how a newspaper discusses a certain issue, regarding the content as well as how many articles are dedicated to the matter. Further elaboration on these aspects will be involved while discussing the actual content of the articles.

When it comes to the length of the articles, both newspapers show on average almost a similar number of words per article. (The ‘Climate Special’ of the PZC (Article 27, PZC5₎ has been disregarded in calculating this average, because it does not represent a general newspaper article or column.) The PZC allocates on average 654 words to articles which mention or discuss the topic of climate change. For DDL the word average per article is at 644 words. However, to be able to judge if the emphasis of an article is actually on climate change, the content need to be studied as well.

2. Article Content

Notwithstanding the foregoing finding that 58 articles mention or discuss climate change, the difference in mentioning (i.e. more as a side issue) and actually discussing the topic (i.e. as -one of- the main issues) is interesting regarding the matter of attention of each regional newspaper towards climate change. To set out this side- versus main issue aspect, the newspaper articles are evaluated by the appearance of ‘climate-‘ (klimaat-) within the ‘head’ and/or ‘lead’ of the article, compared to when this is not the case. The ‘head’ of the article is represented by the title, and the ‘lead’ concerns the short introduction into the article. Not every newspaper article is provided with a lead: this is generally merely applicable to the more elaborate articles. By using the word ‘climate-’ (klimaat-) as criteria, articles mentioning compound words in their head or lead are counted in as well (e.g. climate change (klimaatverandering), climate policy (klimaatbeleid), climate scepticism (klimaatscepsis) etc.). Out of the 58 articles, 27 are mentioning ‘climate-‘ in their head or lead (representing 46.6 percent of the total amount). This suggests that in almost half of the articles climate change related issues are (one of) the main topics discussed.

5_{IN SEQUEL: When reference is made to a regional newspaper’s article, see file 1 or 2 in: “Research Attachments”} https://drive.google.com/drive/folders/1UO4UKWjKMKPH7At6kakWE2mYWrCetD4B

34.4%

63.6%

DDL: Head and/or Lead

mentioning

Climate-Yes (8 articles) No (14 articles)

52.8%

47.2%

PZC: Head and/or Lead

mentioning

Climate-Yes (19 articles) No (17 articles)