Influencers - a non-essential profession in society?: A discourse analysis of BBC News and The Guardian comment fields on Facebook

(1)

Influencers: a non-essential profession in society?

A discourse analysis of BBC News and The Guardian comment fields on Facebook

Influencers: ett icke-väsentligt yrke i samhället?

En diskursanalys av BBC News och The Guardian kommentarsfält på Facebook

Jennifer Frisell & Chanapha Intichan

Faculty of Arts and Social Sciences Media and communication

15 ECTS Credits Didem Özkul 2021-06-04

(2)

Sammanfattning

I dagens medielandskap är gränserna mellan traditionella medier och digitala medier

sammanflätade. Därför använder många nyhetshus sociala medieplattformar som Facebook för att dela sina artiklar för att nå en större publik. Samtidigt skapar det en möjlighet för läsare att interagera med artiklarna genom att lämna en kommentar.

Syftet med denna studie är att undersöka vad Facebook-användare kommenterar inlägg som innehåller länkar till nyhetsartiklar om influenser i förhållande till COVID-19. De utvalda nyhetsartiklarna publiceras på BBC News och The Guardian Facebook-sidor. Vår

forskningsfråga är: Vad kommenterar Facebook-användare på inlägg om influencers i förhållande till COVID-19? Dessutom har vi två arbetsfrågor för att stödja forskningsfrågan; (1) Vad är

likheterna och skillnaderna i svaren när man jämför kommentarer om inlägg från 2020 och 2021? (2) Vad är användarnas åsikter om påverkare i förhållande till COVID-19? För att stödja frågorna använder vi oss av teorierna public sphere och discourse theory.

När det gäller metod genomförs studien med ett kvalitativt tillvägagångssätt, vilket innebär att vi granskade fyra Facebook-inlägg, varav 50 kommentarer från varje inlägg med hjälp av diskursanalys. Valet av inlägg baseras på att artiklarna i inläggen handlar om influencers i samband med COVID-19. Därefter bearbetades materialet i programmet NVivo, där vi kunde identifiera de mest använda orden bland kommentarerna.

Resultaten visar att de teman som kan identifieras i inläggen är influenser som ett yrke, påverkares etiska beteende och samhälle. Sedan analyserar vi teman, några av de mest använda orden och skrivstilarna i olika sammanhang. Till exempel hävdar många kommentatorer att det inte är ett verkligt jobb att vara en influencer, och det är därför absurt att influencers får resa utomlands trots att restriktionerna innebär att UK-invånare bara kan resa utomlands för arbete. I linje med detta kritiserar många också restriktionerna eftersom de endast gäller en specifik grupp, vilka influencers inte faller inom kategorin. Andra aspekter som kan grundas är att influenser kan göra vad de vill och de är skickliga på marknadsföring.

Slutsatsen är att bilderna och rubrikerna fungerar som en bas för att förstå hur kommentatorerna kan uppfatta influencers. Dessutom kan den allmänna opinionen variera beroende på kunskapen hos människor om COVID-19. Vi kan också indikera att Facebook- användare tenderar att lämna kommentarer med beskrivande och övertygande skrivstilar snarare än förklarande.

Nyckelord: Facebook diskurs, online-nyhetskommentarer, diskursanalys

(3)

Abstract

In today's media landscape, the boundaries between traditional media and digital media are intertwined. Thus, many news outlets can use social media platforms such as Facebook to share their articles to reach a larger audience. At the same time, it creates an opportunity for readers to interact with the articles by leaving a comment.

The purpose of this study is to examine what Facebook users comment on posts that contain links to news articles about influences in relation to COVID-19. The selected news articles are published on BBC News and The Guardian Facebook pages. Our research question is: What do Facebook users comment on posts about influencers in relation to COVID-19? Besides, we have two subquestions to support the research question; (1) What are the similarities and differences in the responses when comparing the comments on posts from 2020 and 2021? (2) What are users' opinions about influencers in relation to COVID-19? To support the questions, we use the theories public sphere and discourse theory.

Regarding the method, the study is carried out with a qualitative approach, which means that we examined four Facebook posts, of which 50 comments from each post by using discourse analysis. The choice of posts is based on the articles in the posts being about influencers in connection with COVID-19. Then the material was processed in the program NVivo, where we could identify the most used words among the comments.

The results show that the themes that can be identified in the posts are influencers as a profession, influencers' ethical behaviour and society. Then we analyse the themes,

some of the most used words and writing styles in different contexts. For example, many commentators argue that being an influencer is not a real job, and thus it is absurd that

influencers are allowed to travel abroad even though the restrictions imply that UK residents can only travel abroad for work. In line with this, many also criticise restrictions as they apply only to a specific group, which influencers do not fall into the category. Other aspects that can be founded are that influencers can do whatever they want, and they are skilled at marketing.

We concluded in the study that the pictures and headlines serve as a basis for how the commentators can view influencers. Moreover, public opinion may differs depending on the knowledge of people regarding COVID-19. We can also indicate that Facebook users tend to leave comments with descriptive and convincing writing styles rather than explanatory.

Keywords: Facebook discourse, online news commenting, discourse analysis

(4)

Acknowledgements

The study is co-authored and has been formulated in symbiosis between Jennifer Frisell and Chanapha Intichan. During the work process, some parts are co-written. However, we were also responsible for different parts of the study. Therefore, the chapters were written as follows:

Frisell: 1.0, 1.1, 1.2, 1.3, 1.3.3, 1.4, 1.5, 2.0, 2.1.4, 2.1.5, 2.1.7, 2.1.8, 2.2, 3.0, 3.2, 3.3, 3.5, 3.8, 3.9, 4.0, 4.1, 4.2, 5.0, 5.1.1, 5.2.1, 5.3, 6.0, 6.1.1, 6.1.2, 6.1.3, 7.0, 7.1, 7.2, 8.0

Intichan: 1.0, 1.1, 1.2, 1.3, 1.3.1, 1.3.2, 1.3.4, 1.3.5, 2.0, 2.1, 2.1.1, 2.1.2, 2.1.3, 2.1.5, 2.1.6, 2.2, 2.3, 3.0, 3.1, 3.2, 3.3, 3.4, 3.6, 3.7.1, 3.7.2, 3.7.3, 4.0, 4.3, 4.4, 5.0, 5.1.2, 5.2.2, 5.3, 6.0, 6.1.3, 7.0, 7.1, 8.0

We would like to express our gratitude to Cornelia Brantner for being our supervisor, encouraging and guiding us with her wise counsel during the work process.

(5)

1. Introduction

Various media have long been used to disseminating news in society, such as print media (e.g. newspapers) and the Internet (McQuail, 2010). Moreover, McQuail (2010, p.40) claims that the media can be intertwined, for instance, online news that "are clearly extensions of newspaper journalism, although online news itself is also evolving in new directions with new capabilities of content and new forms". In relation to today's media landscape, online news has moved a step further. Several news outlets do not only distribute news articles on websites but also on social media platforms, such as Facebook.

Facebook is one of the leading social media platforms (Van Dijck, 2013) that offer news outlets the opportunity to create a news page (Facebook, n.d.-a) and allow its users to participate in the page through, for instance, comment. In other words, when news outlets share the news articles with their audience through their news page, they can reach more audiences and receive more feedback.

In terms of news, it is inevitable that since the outbreak of the COVID-19 pandemic at the end of 2019, the journalists have focused on reporting on it since the pandemic affects both individuals and society. This aligns with what Harcup and O'Neill (2001) emphasise that news that concerns relevance in society is news that has value. Therefore, if a news story has value, it can increase reading interest among the newsreaders (McQuail, 2010).

Besides, influencers also have an important role in (digital) society. Many researchers argue, just like Chopra, Avhad & Jaju (2021, p.78), that "an influencer can be anyone, from a fashion blogger on Instagram to a wedding photographer on Pinterest, to a cyber-security expert who tweets on Twitter". That means an influencer is ordinary people that share their lifestyle through social media platforms (Abidin, Barbetta, Lee & Shan Miao 2021; Van Driel & Dumitrica, 2021).

On social media, influencers can gain followers and therefore build their own audience base.

This, in turn, means that influencers can affect their followers in terms of purchase intention or create awareness of other companies (Chopra et al, 2021). While influencers seem to have an important role economically, Abidin et al. (2021) assert that influencers also have a social role, for instance, by disseminating socially important information. With this in mind, it has become clearer when some governments collaborate with influencers during COVID-19 to promote new social norms, such as handwashing and social distancing (Abidin et al., 2021).

Harcup and O'Neill (2001) also mention that influential people (e.g. influencers) have news value. Therefore, by writing an article based on these two news values, the journalists can grab the reader's attention.

(8)

This study aims to examine what users comment on posts containing links to news articles that were published in 2020 and 2021 and are about influencers in relation to COVID-19 on Facebook. More specifically, we will examine the comments from BBC News and The Guardian Facebook pages. By studying Facebook users' comments with discourse analysis, we intend to shed light on the discourses that occur through opinions and how the characteristics of the comments may differ or be the same over the years.

1.1 Background and problem description

At the end of 2019, a new type of coronavirus was discovered in Wuhan, Hubei Province, China (World Health Organization, 2020). Since it was discovered in 2019, the coronavirus is

abbreviated as COVID-19 (World Health Organization, 2020). Therefore, we will use this abbreviation (COVID-19) in this study. As COVID-19 has spread worldwide, the virus was classified as a pandemic by the World Health Organization on March 11, 2020 (World Health Organization, 2020). A pandemic can be considered a crisis in society, and thus it gets a lot of media attention as it has great news value. In line with what Harcup and O’Neill (2001) mention, news stories need to satisfy criteria such as unexpectedness, bad news, and relevance to have news value. This indicates that it is important for journalists to report news stories about COVID-19 since it is a novelty that is relevant to society.

This study will concentrate on the selection of British news outlets, BBC News and The Guardian since the United Kingdom (UK) has introduced many restrictions and lockdown laws during 2020 and 2021. Besides, these news outlets report news in English, which means that they reach a larger audience as, according to Lyons (2021), more than 1.3 billion people speak English and thus, English is the third common language in the world. Additionally, this study is written in English, and therefore there is a lesser risk that the comments may be mistranslated.

The first case of COVID-19 was registered in the UK in January 2020 (Weaver, 2021).

Since then, the number of COVID-19 infected cases is increasing. The data presented by The Visual and Data Journalism team (2021) at BBC News shows that there are over 4.4 million confirmed total infected cases in the UK. Due to the data, Steed and Cavanagh (2021) state that the UK prime minister, Boris Johnson, has announced three national lockdowns during 2020 and 2021 in order to prevent the spread of COVID-19. Citizens are only allowed to leave their homes if necessary, such as travel overseas for work and grocery shopping (Cabinet Office, 2021;

Steed & Cavanagh, 2021).

(9)

Since the outbreak of the COVID-19, the UK government has urged its residents to work from home if possible. For some professions, this statement is difficult to apply in practice. An example is social media influencers, who, according to Van Driel and Dumitrica (2021) can earn money through social media platforms. A way for influencers to earn money is, for instance, through collaboration with companies. Although it seems that influencers can work from home, Van Driel and Dumitrica (2021) claims that influencers have to fulfil the demands of their followers. This means influencers have to have varied content in order to maintain their

followers. A study conducted by Van Driel and Dumitrica (2021, p.78) showed that “more than three-quarters of the sponsored posts were photos taken in a travel-related setting”. In terms of the pandemic, this can be problematic as influencers promote a lifestyle that is difficult for the audience to relate to. Besides, the restrictions made it difficult for influencers to travel abroad and create varied content to maintain their audience. On the other hand, Maares, Banjac and Hanusch (2021) underline that influencers must have a friendly relationship with their audience to sustain the audiences’ attention. In light of COVID-19, this type of relationship can change if influencers do not follow the restrictions. To sum up, it is challenging for influencers to know how to behave during this pandemic as they have to maintain their audience but at the same time be a good role model by following the restrictions. Nevertheless, there are other ways for

influencers to earn an income during the pandemic. Abidin et al. (2021) describe that during COVID-19, some governments collaborate with influencers to spread information.

Facebook was founded in 2004 and has since been a user-generated content platform that allows its users to participate in each other’s everyday lives or communities within the platform (Van Dijck, 2013). In other words, Facebook users can use the platform to debate a certain topic (Domingo et al., 2008). Additionally, Facebook also offers several features for users to interact with the posts, such as liking and commenting. NapoleonCat (2021) presented a table that shows that there is an increase in Facebook users in the UK (see Appendix 1), and Tankovska (2021) claims that Facebook is one of the most leading social media platforms in the UK. Hence, many news outlets such as BBC News and The Guardian use Facebook as an external platform to share their online articles. By doing that, the journalists can reach more audiences. Furthermore, as Facebook offers users the opportunity to exchange dialogue with each other, Harcup and O’Neill (2017) claim that journalists can thereby instantly receive feedback from their audience.

However, the feedback from the audience only has a role to some extent since, after all, the journalist has the power to decide the articles’ content (Domingo et al., 2008).

As stated earlier, journalists are dependent on certain criteria when reporting on news as the news need to have news value; that news is, for instance, news that includes stories relevant

(10)

to society and news of influential people (Harcup and O’Neill, 2001). In relation to COVID-19 as a pandemic, the journalists have to focus on producing news about COVID-19 since that is what the public are concerned about these days. However, to further increase the news value of an article, the journalists can also include influencers in the article. Later on, when the journalists have shared the article on Facebook, it implies that the journalists can influence the opinions of Facebook users who have seen the posts. Hence, people can express their own opinions via comments regarding the news story, which in this case is influencers in relation to COVID-19.

Negative comments about influencers can have a major impact because the influencers can lose followers, be exposed less on social media, and eventually lose collaborations with companies.

This is because all Facebook users can see the comments as the BBC News and The Guardian news pages are publicly accessible. However, there are merely two requirements that users need to have in order to read comments or leave comments: internet access and an account on Facebook.

As mentioned earlier, a new type of coronavirus causes the pandemic, and the pandemic is still part of society, so there is a limited number of previous research, especially about how Facebook users comment on news posts regarding influencers and COVID-19. Hence, it is essential to provide more research on this topic. Furthermore, as the study covers two different periods, the beginning of the pandemic and about one year after the pandemic, there can be a comparison between the periods. In other words, this study will contribute with new insights into what Facebook users are commenting about influencers in relation to COVID-19 during 2020 and 2021.

1.2 Purpose and Research Question

The purpose of this study is to examine what Facebook users comment on posts that contain links to news articles about influencers in relation to COVID-19. As mentioned earlier, the selected news articles were published on BBC News and The Guardian Facebook pages.

Furthermore, since the COVID-19 is an unexplored area in many research fields, we believe that by examining the comments of the posts, we can then fill the gap of understanding the discourse within the comment fields on Facebook during the pandemic. Therefore, this study is based on one main research question (RQ) and two subquestions (SQ).

The main research question is:

RQ: What are Facebook users commenting on posts about influencers in relation to COVID-19?

(11)

The following working questions will help us to answer our research question:

SQ1: What are the similarities and differences in responses when comparing comments on posts from 2020 and 2021?

This first subquestion is used to find if the comments differ or are the same during different times.

SQ2: What are users’ opinions of influencers in relation to COVID-19?

This second subquestion is used to get an idea of how users view influencers based on the content of the post and how it can be in line with the discursive constructions.

1.3 Definitions of central terms

In the sections below, we will define the central terms of this study.

1.3.1 Social media

It is challenging to define social media as there are various types of social media. In her book, Van Dijck (2013) explains that there are four types of social media, which we will only refer to two of them since we find them the most relevant to this study. The first type is social network services (SNSs), which promote interpersonal contact, such as Facebook. The second type refers to user-generated content (UGC) sites. That means users can upload their amateur or

professional content. Two of the most well-known UGC sites to date are Youtube and Wikipedia. Van Dijck (2013) emphasises that these kinds of sites encourage the creativity of users. However, Van Dijck (2013) also points out that defining a specific platform into a particular type is challenging. An example is that Facebook fits in with these types of social media. The platform allows its users to upload personal photos, share posts, and be part of specific communities. In the context of the study, the term social media refers to UGC sites, primarily Facebook, as we are examining users’ comments on Facebook.

1.3.2 Audience

Audience is, according to McQuail (2010, p.398), both “a product of social context (which leads to shared cultural interests, understanding and information needs) and a response to a particular

(12)

pattern of media provision”. In this study, we refer the term to both commentators and spectators of news and influencers.

1.3.3 Influencers

An influencer is a person who has an audience on social media platforms. According to Lee Coco and Eckert (2020), an influencer is a person who is just as ordinary as anybody else. The difference is that an influencer has a large number of followers and that they share their life through text, audio, pictures, and video posts. Commonly, the post attracts an audience that has the same kind of interest, which is what makes these people into online personalities (Lee Coco

& Eckert, 2020). In addition to posting regularly, these online personalities build trust between them and their followers (Lee Coco & Eckert, 2020). This study will be using influencers as an umbrella term because tv-celebrities can also be classified as influencers.

1.3.4 Mass media

In the study, we shed light on the relationship between Facebook posts and Facebook users.

McQuail (2010) claims that mass media can refer to several types of communication channels, such as newspapers (refers as traditional media) and social media. Thus, the term is used as a collective term in the study to describe the interplay between traditional media and social media.

1.3.5 Public Discourse

Keller (2013, p.73) defines the term as “discourse with an orientation to a general audience in a public domain mediated by the mass media”. To apply with this study, the term is used to explain how commentators on Facebook create and maintain discourse in the comment fields.

1.4 Delimitations

The first delimitation we did was to select news outlets that write articles in English since the language is one of the world’s most spoken languages, and thus we believe that there is an opportunity for more debate on Facebook. In addition, our study is written in English, which means that we do not have to translate the materials and hence there is no risk of

(13)

misinterpretation of data. Afterwards, we decided to focus on the news outlets in the UK since we are more familiar with the UK news outlets than news outlets in the United States. Finally, speaking of English, the influencers are based in the UK and communicate with their audience in English; therefore, they can reach out to many people, and more people on the news pages may recognise them.

This study is delimited to two of the British news outlets: BBC News and The Guardian, since, during the pandemic, the UK has introduced many restrictions, for instance, lockdown laws. Besides, the articles are written in English and reach a larger audience worldwide. BBC News (n.d.) has roughly 56 million followers on Facebook, while The Guardian (n.d.-a) has 8,6 million followers. BBC News is part of The BBC, which is the world’s largest public service broadcaster based in the UK (BBC, n.d.). Thus, their mission is to deliver current news to the people worldwide (BBC, n.d.). UK households pay the license fee to fund The BBC (BBC, n.d.).

On the contrary, The international news organisation Guardian Media Group owns The Guardian, and the Scott Trust is the only shareholder (The Guardian, n.d.-b). The Guardian (n.d.-b) claims that their readers voluntarily funded the news outlet. This means it is optional for the newsreaders to decide if they want to pay the desired amount to support the news outlet or not.

After setting the news outlets’ delimitation, we selected two posts from each news outlet Facebook page. The choice of posts is based on the fact that the articles in the posts have to contain news about influencers in relation to COVID-19. Furthermore, to answer our RQ: What are Facebook users commenting on posts about influencers in relation to COVID-19? We selected the comments of the posts published in 2020 (the beginning of COVID-19) and 2021 (roughly a year of COVID-19).

All of the delimitations above will be discussed in line with discourse analysis to understand the research area comprehensively. We will discuss the delimitations further in chapter 3.5, Procedure and Selection.

1.5 Outline

The first chapter begins with the introduction, which gives an insight into the study with a background of the pandemic and the problem description of the study, which connect with the purpose and research question. This is followed by the chapters about definitions and

delimitations. In chapter two, we present the theories, theoretical frameworks and previous research that are relevant to the study. In chapter three, there is a description of the first stage,

(14)

the second stage, the chosen approach (discourse analysis), and data collection. This chapter also discusses ethical aspects and reliability, validity and generalisation of the study. In chapter four, the results and the analysis are presented. Moreover, we divide the analysis into two parts:

pictures and headlines and the comments. In chapter five, we discuss the results with support from theories, theoretical frameworks, previous research and the first stage. Chapter six begins with a conclusion of the results. The chapter also provides answers to subquestions and research question. In chapter seven, there are suggestions for future study and descriptions of limitations.

Finally, the study finishes in chapter eight with implications for society. There is a separate outline for each chapter, which is written under every chapter heading.

(15)

2. Theories, theoretical frameworks and previous research

This chapter presents theories, theoretical frameworks and previous research that support the study's purpose and research question. We begin the chapter by presenting the theory public sphere to be able to dive deep into the public opinion and online (news) discussions to create a comprehensive understanding of the discourse theory. Then we explain the role and significance of influencers in the media landscape and how influencers are considered to have news value.

Later, we review two previous research that can provide support for the study topic. Finally, we will summarise theories, theoretical frameworks, and previous research to account for what these literature reviews may contribute to increasing the theoretical relevance and understanding of this research.

2.1 Theories and theoretical frameworks

2.1.1 Public sphere

Habermas (1962/1989) described that the public sphere refers to a social space in which individuals may engage and thereby contribute to the formation of a democratic society.

Furthermore, the public sphere refers to a space for conversations where "everyone" can participate, and thus, public opinion can be created (McQuail, 2010; Lindgren, 2017). In other words, a public sphere is an area where individuals can talk, discuss, and debate. However, although the sphere may seem inclusive for everyone, it does not mean that everyone can have access to it.

Before the 1700s, those in power were usually considered to join public debates, such as a king or lord, while people in other social classes were private spectators (Lindgren, 2017).

However, after that, the combination of the rise of capitalism and the emergence of new cultural forms in art and literature enables individuals to debate in, for example, coffee houses or salons (Habermas 1962/1989; Lindgren, 2017). Habermas (1962/1989, p.33) states that "the coffee house not merely made access to the relevant circles less formal and easier; it embraced the wider strata of the middle-class including craftsmen and shopkeepers". Although it seems more easily accessible to join the debates, Lindgren (2017) believes that there are other factors that can affect whether someone can participate in the conversation, such as education, race, and gender.

In relation to today's digitalisation, the public sphere has shifted from coffee houses and salons to social media platforms (e.g. Facebook). Therefore, there are new rules and norms that individuals must take into account. Mass media, such as newspapers and social media, serve as

(16)

the basis for constituting the public sphere. Moreover, according to Habermas (1962/1989), the media is powerful because it can manipulate the audience instead of creating a public sphere. In line with his study, when a journalist publishes an article on Facebook, the journalist can set a specific agenda, which may later influence audiences' opinions.

Regarding social media, individuals can gather regardless of place and time, and therefore there are constant topics that individuals can discuss on different social media platforms. Thus, it leads to a new concept called digital public sphere. Papacharissi (2002, p.11) claims that "the internet could facilitate discussion that promotes a democratic exchange of ideas and opinions". In social media platforms, such as Facebook, people can express their opinions. Additionally, other users can also interact with the posts by using several interactive features, for instance, comments, likes, and emojis.

2.1.2 Public opinion

As previously mentioned, a public sphere enables the exchange of dialogue, which means that public opinion can then be formed. But what is public opinion? In 1965, Harwood Childs published a book where he defines public opinion into 50 definitions (Noelle-Neumann, 1995).

However, Noelle-Neumann (1995, p.34) developed the definitions and asserted that the definitions could be split into two concepts as following

● “Public opinion as rationality: it is instrumental in the process of opinion formation and decision making in a democracy”.

● “Public opinion as social control: its role is to promote social integration and to insure that there is a sufficient level of consensus on which actions and decisions may be based.”

While the concepts above shed light on the term public opinion, Davison (1968, p.188) underlines that “there is no generally accepted definition of the term”. However, one cannot conclude that the term is meaningless (Price, 1992). For several years, many researchers have tried to define the term, such as Peters (1995, p.3), who points out that public opinion represents individuals’

opinions as a group, and it is a “clear and direct utterance from the citizenry.” Given this, McGregor (2019) describes that public opinion is the foundation of democracy as it is a collectively constructed representation of the public.

(17)

Despite the fact that public opinion remains one of the most ambiguous concepts in the social sciences, it is receiving growing focus from scholars across a wide range of fields and disciplines (Donsbach & Traugott, 2008), especially when social media networks like Facebook become more relevant in daily life (McGregor, 2019). Facebook offers individuals a digital sphere to express their opinions towards a post through, for example, comments and emojis. However, in relation to mass media, Peters (1995) argues in his study that the problem of public opinion is that the mass media merely provides their audiences with a view of the public sphere but not the opportunity for the audiences to participate in it. Therefore, it is clear that the public sphere has an influence on how public opinion is constructed.

2.1.3 Online (news) discussion

Ziegele, Breiner, and Quiring (2014) conducted 25 interviews and analysed 1580 users’

comments. Through that, they could identify in their study that there are twelve discussion factors of why users respond to the comments of others: aggression, controversy, facticity, unexpectedness, negativity, personalisation, simplification, comprehensibility, uncertainty, humour, formal descriptors, and discourse architecture. Ziegele et al. (2014) continue to describe that uncertainty, in this case, seems to activate a reader's knowledge, and therefore, this kind of comment can receive more feedback and, thus, it gives rise to discussion. These discussion factors are the basis for why users discuss a certain topic on news websites and news pages on Facebook (Ziegele et al., 2014). As mentioned above, the definitions of the term public opinion are complex. However, several researchers such as Peters (1995) and McGregor (2019) agree that the formation of public opinion is dependent on individuals who are willing to express their feelings and opinions; therefore, it can represent the opinion of a group of individuals. The shift from salons or coffee houses has moved to a digital sphere like Facebook. On Facebook, users can follow a news page by clicking on the follow button on the page, which is equal to joining an online community. This is due to an online community consisting of individuals that have similar interests (Lindgren, 2017). In this case, it means that people who choose to follow the news page have the same interest: to receive news articles in their personal Facebook field. In connection with digitalisation, in other words, public opinion is socially constructed through, for instance comments that the users write, and therefore the comments can be investigated contextually (McGregor, 2019). Furthermore, on Facebook, users can also express their opinions through visual symbols, which can also strengthen their expression of the comment. Kabadayi and Price

(18)

(2014) concluded in their study that the popularity and contribution of brands would be limited if they did not receive comments and likes.

2.1.4 Discourse as a theory

In terms of theory, discourse is a broad theory that has expanded enormously in the field of social science (Keller, 2013), and therefore there are several definitions of the theory. Keller (2013, p. 2) defines the theory as follows

“In social science there is a basic agreement that the relationship between human beings and the world are mediated by means of collectively created symbolic meanings systems or orders of knowledge.”

It means that knowledge can be found just by looking at the relationship created between the human being and the world, which can then create a symbolic meaning to get the knowledge.

The term “discourse” can be explained in English as a simple conversation where there is an interchange between different people, while in other Romance languages, it is the usual term, for example, “learnéd speech”, a lecture and a presentation (Keller, 2013). Discourse is often

constituted in relation to signs of underlying structural patterns or rules that are part of the production of meaning (Keller, 2013). Commonly in the English language, the term discourse has a different introduction; it demonstrates an expansion of different discourse-related perspectives that can be used when doing an analysis (Keller, 2013). According to Keller (2013), discourse is something we use to gather more information and understand the knowledge and the social ensemble.

Jørgensen and Phillips (2002) argue that discourse theory can be defined as the general idea of how language is structured in different contexts when people take part in different social lifestyles. Given this, Talbot (2007) underlines that text can be in verbal or written form performed by one or more people, and it constitutes discourse. In other words, discourse is a cultural activity (Talbot, 2007). However, in terms of online discourse, all internet users have the same conditions to express their opinion, but only a few people dominate the discourse

(Papacharissi, 2002).

For this reason, one may wonder why is it important to define discourse as a theory?

Jørgensen and Phillips (2002) explain that discourse theory is helpful when analyzing all kinds of social phenomena since the theory aims to understand various discursive constructions of the social. Along with the arguments above, Keller (2013) agrees that discourse theory is a general

(19)

theoretical view on the linguistic constitution of meaningfulness of reality. Things that can be considered to constitute discourse are, for example, different social practices that use written or spoken language as well as other symbolic forms that can be seen in social practices. The practical use of signs and meanings of phenomena, individual interpretations can be understood as parts of a more comprehensive discourse and symbolic orders and actions that can be

reconstructed (Keller, 2013). Keller (2013) continues to explain that individual interpretations can be understood as part of a more comprehensive discourse structure that can be temporarily produced. Furthermore, social phenomena are never finished or total, and this is the overall idea of discourse theory which means that the meaning can change and is never a constant thing (Jørgensen & Phillips, 2002).

The definitions proposed by several researchers above provide insight into how discourse can be understood as a theory. Moreover, discourse theory is the foundation of our analytical work as we aim to examine what Facebook users comment on posts that contain links to news articles about influencers in relation to COVID-19.

2.1.5 Online News Engagement

A study by Chen and Pain (2019) suggests two dimensions of online news engagement: Content- Interaction Engagement and Exposure Engagement. Content-Interaction Engagement implies that there are several ways for users to engage with online news, for instance, rating or sharing news, commenting, or creating content, and they can also prove their loyalty toward online news brands (Chen & Pain, 2019; Krebs & Lischka, 2019). While Exposure Engagement refers to attention, the enjoyment that occurs when reading news on Facebook and the frequency of receiving news via social media (Chen and Pain, 2019).

Krebs and Lischka (2019) highlight that online engagement is important for brands to create trust and loyalty toward the news brands, and it also creates a growing relationship. However, it does not necessarily mean that commenting on news brands will be good for the news value.

According to Krebs and Lischka (2019), user engagement with news brands can often be seen as something more positive for the audience itself compared to the news brand. However, if the brand awareness is high, it is also creating a chance that the audience will be able to recognise it, which is important when brands are facing growing competition (Krebs & Lischka, 2019).

Furthermore, Krebs and Lischka (2019) underline in their study that users tend to seek

information more passively by only reading; however, commenting is still occurring and is mainly driven by social-interaction motives.

(20)

To further develop this framework, Möller, van de Velde, Merten and Puschmann (2020) propose three modes of news engagement online:

● Routine mode (people consume news as a part of their daily life)

● Search mode (people search for recent and relevant news)

● Social media mode (people encounter with news on social media)

Since social media is used as a channel to expose the news, users who have a low political interest will probably only read the headlines and teasers (Möller et al., 2020).

2.1.6 Online news commenting

Lee and Jang (2010) define comments as personal opinions where individuals can express themselves freely. Furthermore, comment-writing does not only mean that individuals can express their opinions about certain news topics (Lee & Jang, 2010); a comment also functions as feedback to journalists (Kim, Lewis & Watson, 2018).

Facebook is one of the platforms that several news outlets use to link and share their articles online (Rowe, 2015), and the platform also allows its users to comment on a news post directly. That means the journalists can directly get feedback from the newsreaders, which may help them understand their audience’s interests (Winter, Brückner & Krämer, 2015). However, although Facebook allows the commentators to interact with the journalists through the posts easily, Diakopoulos & Naaman (2011) conclude in their survey study that users would like to receive more responses from journalists to clarify or answer questions. Interestingly, this is in line with what Hille and Bakker (2014, p.570) discuss in their study that there are “no examples of interaction between users and journalists”.

Another reason why several news outlets use Facebook is that they can reach more audiences. A report shows that in March 2021, Facebook had 56,3 million users in the UK; this is an increase of almost 11.5 million more users since the counting from March 2020 (Tankovska, 2021).

In line with the earlier section on discourse, online news commenting is relevant due to the increasing prevalence of leaving and reading comments online (Kim et al., 2018). However, in light of online commenting, Lindgren (2017) claims that there is a phenomenon called trolling on social media, which means that trolls disseminate disinformation and reinforce strong emotions and reactions among the users. In other words, trolls can manipulate people’s opinions and affect public opinion as a whole. While Lindgren (2017) asserts that trolls disseminate

(21)

disinformation and can affect public opinion online, a study conducted by Kim et al. (2018) that compared comments on news outlets’ websites and Facebook shows that trolls were less

common among Facebook users than on news outlets’ websites. However, the number of trolls did not result in a better quality of the discourse since Facebook commenters on Facebook did not think their peers were more reasonable, knowledgeable or less offensive compared to the commenters on news outlets’ websites (Kim et al., 2018).

Another aspect of online news commenting that should be discussed is anonymity.

Facebook is used as a platform to connect with friends and family. News commentators can take their network of contacts into account when commenting on posts since the Facebook news feed feature can reveal any activity that other users in the same “network” engage in via their Facebook account (Hille & Bakker; 2014; Rowe, 2015; Kim et al., 2018). Besides, it is also possible to see users’ usernames and pictures on Facebook comment fields, which results in users not being anonymous (Winter et al., 2015). Furthermore, Facebook does not allow pseudonymous usernames (Facebook, n.d.-b). The combination of visible activity among the network and the real name policy makes it difficult to be anonymous on Facebook. As a result of the lack of anonymity, the commentators on Facebook are less likely to leave an abusive or offensive comment (Hille & Bakker, 2014). Besides, Winter et al. (2015) state that using a public profile may increase a person’s credibility.

2.1.7 Influencers' role and significance in the media landscape

With the increased use of the internet, there is now another way to influence people, which is through influencers. Influencers, private companies and governments are a few of the actors that use social networks to pursue their goals (Galeotti & Goyal, 2009). The question is how the influencers can influence their audience and what is the strategy for it. For example, social interactions are used to influence actions, such as purchasing a product recommended by an influencer (Galeotti & Goyal, 2009). Especially, the younger audience is the one that different brands want to reach out to because they are the generation that is using ratings or are

commenting and sharing their content (Krebs & Lischaka, 2017). This is important because that will help the brand to reach out to even more people if the audience decides to, for instance, comment or share the post. With the help of technology, there is a higher possibility that we can now reach out to a much larger audience with our content and distribute our products (Maares, et al., 2021). With a larger audience, the influencers can easily be used to distribute different products for brands and then get financially paid for their work (Maares et al., 2021). According

(22)

to Maares et al. (2021), influencers work in a constant competition where they want to get the audience's attention to be able to continue with their passion. To be able to keep the audience, influencers have to keep up with the connection to the audience by continuously sharing personal information to maintain a friendly relationship with them (Maares et al., 2021)

With influencers being a great possibility to market items and services, it is then important to get an insight into how the audience who are following influencers think about them and the marketing. Before social media is dominated in society, marketing takes place via television, radio or print media (e.g. magazines). However, with the increasing use of the internet, companies need to create an online identity, and they also collaborate with influencers to create long-term relationships with their target groups (Lee Coco & Eckert, 2020). In fact, it has become more common for companies to use influencers for marketing their product; the lines of

communications are then getting more blurred between the interpersonal and commercial communication between the influencer and their audience (Lee Coco & Eckert, 2020). As Lee Coco and Eckert (2020) discussed, the relationship between an influencer and their audience has a stronger connection since their followers can interact with the influencer, which creates

credibility and attractiveness to the product that the influencers choose to promote. Moreover, influencers are commonly used for advertising products, and the public relations professionals expect the influencer's content to be of the more traditional way of advertising which might not be the case (Lee Coco & Eckert, 2020) because they might choose to advertise it in a more personal way.

In Lee Coco and Eckert's (2020) study, where they interviewed 15 women about why they followed influencers; they commonly said that it was because of the topics that the influencers are addressing, which is for that reason, some of them would choose to click on that person's profile. Another reason for following a certain influencer is if the audience can relate to the influencer's lifestyle, political view, or physical trait (Lee Coco & Eckert, 2020). The study also indicates that the woman that got questioned most likely has a more interpersonal relationship with the influencers. This implies that the audience has a personal relationship with the influencers, and therefore, influencers can market products or services.

2.1.8 News value on SNS and influencers

Some of the questions that arise during the study are what is news, and why do influencers have news value? News value is important to understand in order to get a grip of why news outlets choose to post their articles on social networking services (SNS). Al-Rawi (2017) points out that

(23)

it is important to make studies of how the news selection process is made. In addition, how countries and political actors are linked to net worthiness of the relevance of proximity, ideology, and social significance.

A list that is introduced by Harcup and O’Neill (2001) is showing the criteria that can create news value: the criteria can be seen in the following list

● Celebrity - well-known individuals

● The power elite - for example, influential people, institutions or organisations

● Relevance - content that is socially relevant and that the audience find interesting

What can be seen here is that influencers are influential people and can therefore have news value to some extent. With the increase of the pandemic COVID-19, the media is also increasing their coverage, and since it is increasing, the number of articles written about influencers in relation to COVID-19 are also getting more discussed (Abidin et al., 2021). That is why there is a relationship between influencers and COVID-19.

Facebook is a good social networking service (SNS) because of the allowance of reading and commenting on news articles, and this is because the audience consumes more news from the Facebook-recommended pages than compared to Twitter (Al Rawi, 2017). When looking at the American audience, there is a higher percentage of users that get their news on Facebook’s platform (30 %) compared to YouTube (10 %) and Twitter (8 %) (Al-Rawi, 2017). To attract the audience to read the articles, the journalists need to decide which article to post on a SNS site, and it can, for example, be about certain political, religious or cultural beliefs and values (Al- Rawi, 2017). This implies that newsworthiness gets higher depending on how much it shares the same ideologies and cultural values of the readers (Al-Rawi, 2017).

Moreover, Facebook is considered to be a (new) way of marketing and distributing content in order to reach more audiences since, according to Al-Rawi (2017), it is common to use these posts on Facebook to publish hyperlinks that can direct the traffic from SNS sites to their news website. However, it is also important to note that it can occur that the audience sometimes are not that engaged with the post and only read the headline rather than reading the entire story, which can lead to misunderstandings of the content (Al-Rawi, 2017).

(24)

2.2 Previous Research

The first research investigated users’ comments on news pages and social media. The second research is about how the news articles are writing about influencers in relation to the pandemic of COVID-19.

In an earlier study by Kalogeropoulos, Negredo, Picone, and Nielsen (2017), the authors

performed a cross-national comparative study where the focus lay in comparing the sharing and commenting on a news story on news websites and social media. In other words, the goal of the study is to examine the patterns of news consumption in six countries: Denmark, Germany, Italy, Spain, the United Kingdom, and the United States. Kalogeropoulos et al. (2017) described their analysis which is based on four hypotheses in which one of them refers to hard news (e.g., economic news) and soft news (e.g., entertainment, lifestyle news). The used method was a survey based on data from the Reuters Institute Digital News Report presented in 2016 and various variables, such as gender, age, education and income. The survey was conducted from the end of January 2016 to the beginning of February 2016.

The result showed that 14% of the 2 024 respondents from the UK report that they comment on news posts on social media. That is six percentage points higher compared to the number of respondents who said that they comment on news websites. Furthermore, Kalogeropoulos et al.

(2017) indicate in their study that people who use several social media platforms for different purposes, including news reading, are more likely to leave comments on news websites and share news further, for instance, via emails. Besides, those who are interested in hard news tend to share news via social media and can leave a comment on both social media and news websites.

Although our study seems to consist of soft news articles as it concerns influencers, we find that the COVID-19 pandemic is important news that can affect both socially and economically.

Therefore, we find this study relevant to apply later in the chapter 5 discussion.

The second previous research is written by Abidin et al. (2021); their research concerns media coverage on COVID-19 in relation to influencers. Abidin et al. (2021) surveyed 150 news articles published between January and May 2020 in Australia, China, Japan and South Korea.

Furthermore, the authors examined what journalists were writing about influencers and could therefore code the following categories: income loss, backlash, COVID-19 campaign,

misinformation and influencer strategy. In their analysis, they could state that the way how the articles were written regarding influencers differ depending on local contexts and how the pandemic has affected the country. To note this, Abidin et al. (2021) mention that a common

(25)

thread in the four countries is that all of the articles write about influencers in relation to covid- 19 precisely because it can increase the click-through rate on their articles. In addition, people search out key opinion leaders for knowledge and advice during a pandemic like COVID-19 because there is confusion within the social groups. As a result, the authors emphasise the importance of partnerships between public actors such as governments and health organisations and influencers in aiding COVID-19 recovery efforts and fostering global solidarity. However, the study also indicates that influencers can spread misinformation, such as health advice and vaccinations. Therefore, the fact that influencers can spread misinformation is relevant to highlight as our study intends to examine how newsreaders perceive influencers through their comments on posted articles on Facebook. Abidin et al. (2021) also concluded that when newspapers choose to write about influencers, it gives the impression to readers that influencers play a larger role and have greater importance to the public. In other words, when there is an exchange between the newspapers and influencers, the newspapers get content for their articles, and influencers gain increased awareness among people.

2.3 Summary of the theories, theoretical frameworks and previous research

Throughout the literature, the significant finding is that there are several factors why users engage with Facebook news posts. One of the factors is affiliation with others. Facebook is a social media platform where its users can join communities and comment on posts about certain topics within the communities (e.g. news posts). Hence, public opinion can be formed in each Facebook post. To further deepen the understanding of users’ discussions, the study also mentioned discourse theory. The discourse theory is used to identify patterns by examining the text in both verbal and written form. To put the theory in relevance to the study, the theory is used to understand text in a social context. Moreover, several researchers such as Jørgensen &

Phillips (2002) and Keller (2013) argue that discourse constantly changes the meaning, and thus interpretation may differ. However, it is also important to keep in mind that only a few people that control the discourse on the internet (Papacharissi, 2002).

Rowe (2015) underlines that news outlets chose to use Facebook as a platform to promote their articles. That means it is easier for the users to find and engage with the articles through, for instance, comment. There are several advantages of online commenting as it is a way for

Facebook users to exchange dialogue with each other and a way for the newsreaders to communicate with the journalists (Lee & Jang, 2010; Kim et al, 2018). Through previous

(26)

research conducted by Kalogeropoulos et al. (2017), we can see that news consumption patterns differ depending on countries. The respondents from the UK tend to leave more comments on news pages than news websites. However, a disadvantage of online commenting is that there are trolls, and they disseminate disinformation within the comment fields (Lindgren, 2017).

It is important that the articles must have news value in order to arouse the interests of newsreaders. Harcup and O’Neill (2001) indicate that articles that have news value are, for instance, articles that concern socially relevant content or influential people. The combination of the previously mentioned theories and theoretical works enables a good opportunity to

understand how these can support the research question and hence achieve the purpose of the study by examining users’ comments.

(27)

3. Method

This chapter consists of a description of the qualitative approach, which is discourse analysis. In the following sections, we will give a more profound explanation of the procedure and selection of the material and how the analysis is conducted. The chapter then concludes with a discussion of ethical research aspects and a discussion of validity, reliability and generalisations.

3.1 Research methodology

As mentioned earlier in chapter 1.2, the purpose of the study is to analyse the comments in the selected posts on BBC News and The Guardian news pages on Facebook. We found that a qualitative method, discourse analysis, is suitable to apply in this study because several

researchers such as Smith and Bell (2007) and Taylor (2013) agree that discourse analysis is used as a tool to study language in particular situations. Besides, to get a comprehensive understanding of the study, we also use discourse analysis to analyse the pictures of the posts.

Users have to see the posts before leaving a comment; therefore, we propose that by analysing the post as a whole (including pictures and headlines), we can understand how the picture and headline of each post is represented to the audiences and how it may affect the public’s opinion.

This study does not aim to generalise the findings. It instead aims to analyse the users’ comments in detail.

3.2 First stage

Before we explain our analysis process, we would like to present the first stage. In this first stage, we had analysed the same articles as we examined in this study. Our research question was, “how do BBC News and The Guardian portray influencers in relation to COVID-19 in their online articles?” After we have formulated our research question, we read through the articles and could thus identify seven categories: income, activities, content, restrictions, law, travelling, and professional duty.

Afterwards, we decided to use a qualitative approach: rhetorical analysis as the method.

Rhetorical analysis is suitable since it allows you to divide a text into smaller pieces where you examine how they are connected with each other to get a result. In line with what Batura, Bakiyeva and Charintseva (2020) emphasise when dividing a text, you get rhetorical fragments that you then try to see a relation between the connection and then create a meaning of the text.

(28)

When the method was set, we divided the analysis into two parts based on the categories:

the effect of the pandemic on influencers (which consists of income, content, and restrictions) and influencers’ views and reactions to the pandemic (which consists of travelling and

professional duty). However, we only chose to analyse five of seven categories as we found that some of them are intertwined with each other.

We approached the text by dividing it into smaller parts such as headline and content to analyse it from that. The results showed that the articles from 2020 emphasise how the pandemic affects influencers both professionally and economically. Additionally, we also quote specific sentences to strengthen our analysis further. Some examples are that according to an article published in April in BBC News, the journalist pointed negatively that influencers lose their

“glossy lifestyles” due to COVID-19. The author of The Guardian highlighted in an article from May 2020 that “2020 should have been the year for influencers to get back to their followers’ desire for

“realness””.

On the other hand, the articles from 2021 are instead about influencers who choose to travel during the pandemic. Consequently, it creates a bad image for them and their branding partners. People who follow these influencers on social media also criticise the influencers as they decide to travel like a non-ongoing pandemic. The journalist of the article published in 2021 in BBC News writes that “some commenters accused celebrities of being out of touch with their fans or trying to exploit loopholes in the rules”. The journalist indicated that influencers receive negative responses regarding their travelling during the pandemic. Differing from the article published in 2021 in The Guardian, the journalist mentions that in the influencers’ views: working is not the same thing as being on holiday, even if they are abroad. We could conclude in this stage that the news articles from both news outlets focus on influencers as a professional, and thus they can use both positive and negative aspects to convey about the profession to the public. Regarding influencers that travel abroad for work, both articles, however, question whether the influencers’ travels were necessary.

3.3 Second stage

This study is counted as the second stage since it focuses on conducting discourse analysis of the posts: both the pictures and the comments. The analysis of pictures is based on Gee’s (2010) analytical way, which means that a picture can be divided into several elements (see more in chapter 3.5 Discourse analysis of images). By analysing the most important elements of a picture, we can see what meaning of a picture can contribute to the post as a whole. A comment consists of

(29)

text. Therefore when we analyse a text, we study the different elements. In section 3.7.3, we had presented how the analysis will be carried out in detail. In addition, there is also an illustration of the comment analysis process in section 10.2, appendix b, which provides examples of how the analysis has been conducted.

3.4 Discourse analysis

This qualitative research method of discourse analysis is employed as a guideline to investigate a collection of selected comments in relation to the public sphere on Facebook. It is difficult to have a clear definition of the method as it refers to a number of approaches within several research fields (Smith & Bell, 2007; Taylor, 2013). However, in order to get a grasp on how the analysis can be carried out, Smith and Bell (2007, p.78) state that

“Discourse analysis involves a close examination of text, including visual imagery and sound as well as spoken or written language. It is concerned with both the form of the text and its use in social context, its construction, distribution and reception. It aims to understand and elucidate the meanings and social significance of the text.”

The quote above is in line with what Keller (2013) emphasizes that discourse analysis is an interpretation of a text. Additionally, McQuail (2010) also points out that when a text is constructed, there is connotation and denotation by those who read and those who formulate the text. Taylor (2013, p.53-57) presents several benefits of discourse analysis, which are, for instance as the following

1. “To work with interesting data” – as discourse analysis examines the intrinsic interest of the material, it can give researchers news insights.

2. “To investigate the practices and interactions of ordinary social life” – this means that using discourse analysis enables scholars to explore social phenomena.

3. “To explore people’s opinions” – discourse analysis investigates ideas that are kept in common through culture or within specific social groups or contexts, but it does not examine or make assumptions regarding individual opinions.

Although there are many useful areas in discourse analysis, Schegloff (1997) discusses in the conclusion of his study that discourse is often rendered subservient to contexts placed on it by analysts rather than by its participants. In other words, analysts are likely to interpret the material

(30)

based on their own experience. Another criticism is that words can have different meanings in different contexts where they can incorporate gestures (i.e. nod of the head) or when they are used with a certain sound even if it does not count as a word (i.e. Mm) (Taylor, 2013). Therefore, it can be challenging to examine texts that include an expression like “Mm”.

Despite these criticisms, discourse analysis is still relevant in examining the meaning of words in different social contexts. This is due to its possibility to interpret discourses that are widely in society or within specific social contexts or groups (Taylor, 2013).

3.5 Discourse analysis of pictures

Gee (2010) underlines that the term multimodal texts are used to describe when combining words and images (e.g. pictures and diagrams). Moreover, in terms of discourse, when individuals communicate with each other, they use another modality than just language (Gee, 2010). In this study, we analyse both words and pictures. Therefore, we find it suitable to lay our analytical tool of pictures on Gee’s study. According to Gee (2010), to be able to analyse pictures, one has to identify elements that composed a picture. Then, select some elements that will be analysed.

Gee (2010) gives a few examples that could be elements which are colours, shapes and in some cases, it could be objects in the image. With the help of Gee’s (2010) way of analysing a picture, we will be looking at the picture of each post and identifying the most important element and then interpreting what the element can contribute to the picture.

3.6 Procedure and Selection

In this study, the focus lies on how Facebook users comment on the news articles that have been published on Facebook. Opgenhaffen & Welberg (2018) discussed in their research that

Facebook is a good source for news outlets to publish their articles in order to gain more clicks to their homepage. The news articles were published on two major news outlets in the UK, BBC News and The Guardian. The article selection process was that we first went to the BBC News and The Guardian Facebook page. Then we used the keyword “influencers” in their search query as Harcup and O’Neill (2001), and Abidin et al. (2021) underline that influencers are influential persons, and therefore the news regarding influencers can receive many interactions from the newsreaders. This gave us several articles that involved influencers in it, and from that, we read through the articles to find the ones that talked about influencers in relation to COVID-19.

(31)

Additionally, the articles were selected between 2020 to 2021; therefore, we could compare the empirical data.

As we mentioned in section 1.3.3, the definition of the term influencers is broad.

Consequently, another delimitation we did was that influencers need to be either a TV celebrity or a Youtuber who has their audience on social media. When we selected the articles, we found that the number of comments are up to 3 000 comments (including replies), so we, later on, chose 50 comments from each post, which we believe is a sufficient amount to analyse

profoundly and to avoid the replies as in this study we were interested in the first opinion of the Facebook users without influence from other users. We use our personal Facebook account to collect the comments from the news pages. As the news pages are public, we can see their posts without joining the community.

While extracting the comments from the posts, we discovered that some comments contain a meme. As our analysis is primarily based on discourse (text), we decided that if a comment is a meme that contains a text, we write it manually in an excel file; otherwise, the comment will be excluded.

3.7 Data collection

3.7.1 Collecting the comments

The first stage of our data analysis was to select comments from Facebook posts, which, as mentioned earlier, we chose to view all comments, including potential spam. According to Facebook (n.d.-c), all comments imply that the relevant comments are shown first, and relevant comments are comments from friends and the most engaging comments. We find this option the most suitable as; otherwise, we have to select to see the newest comments first, or the order of the comments may be influenced by our personal network, which is not the intention of this study. Later on, due to the API restrictions of Facebook, we had to manually extract 50

comments from each Facebook post by copying the comments to Excel. Additionally, we did not include replies to comments in this study because we want to investigate the discourse of the comments that are directly commented on the posts and exclude the comments that have been made as a follow from another person's comment. This means that we will be focusing on the primary comments. We set the delimitation of 50 comments since we believe that it gives us enough material to investigate the subject profoundly. To note this, we collected the comments on 6 May 2021. In the section above, we explain our approach to how to investigate the

Influencers - a non-essential profession in society?: A discourse analysis of BBC News and The Guardian comment fields on Facebook