Health crisis communication in Finnish news media: Evaluative images of the Covid-19 pandemic in digital news headlines

When readers mainly browse news feeds instead of reading articles in-depth, digital news headlines have a significant role in creating certain types of images on current events (Mustafa-Awad & Kirner-Ludwig, 2017). These images have the power to shape our perceptions and even influence our behaviours (Mustafa-Awad & Kirner-Ludwig, 2017); thus, the study of news headlines is essential. The Covid-19 pandemic was not only an extensively communicated and mediatised health crisis but also one of the most challenging situations for news media to communicate in the course of the infodemic (Johansson et al., 2023; Lewis et al., 2021). Although other global health crises have occurred previously, the preconditions for the crisis communication of Covid-19 were more complex (Johansson et al., 2023). News addressing the pandemic dominated news coverage in an extreme way, and updated content was produced around the clock (Johansson et al., 2023; see also Ytre-Arne & Moe, 2021). There was even a time when the pandemic almost overshadowed other newsworthy topics of political conflicts and natural disasters in the news headlines (Ytre-Arne & Moe, 2021). On the other hand, the constant flow of information supported the citizens in understanding the bigger picture or, in other words, the social aspect of the health crisis beyond medical perspectives (Quandt & Wahl-Jorgensen, 2022). By early 2023, the pandemic was no longer considered a global health emergency and no longer dominated news media coverage. The focus of the news media in Western countries shifted from the health crisis to the political crisis caused by Russia’s war in Ukraine. However, the need to study and understand the nature of communication during the pandemic, and crises in general, remains when preparing communication for future crises.

The practices of news are often contradictory: The news media must communicate accurate information to citizens but, at the same time, attract readership to maximise newsworthiness, increase advertisement-based revenues, and improve social media shares (Glik, 2007; Jaakkola, 2018; Kilgo & Sinta, 2016; Kivioja, 2018). For example, sensational and exaggerated language deployed in the headlines is a typical way of generating more clicks (Kilgo & Sinta, 2016). Even though crisis communication needs to remain coherent, it must also be appealing when providing information to the public on what is happening and why, and how to behave (Boin et al., 2016). In this article, we consider that the divergent practices of news media offer a fruitful starting point for examining images of the pandemic in the headlines of crisis communication. We enter the realm of crisis communication and news headlines from a linguistic perspective of evaluation. This allows us to analyse the attitudes and emotions in the language used to address the pandemic. Especially, we consider that when evaluations are attached to specific themes related to the pandemic, we can identify unified entities, which we call evaluative images. The linguistic approach complements research in media and communication studies by providing detailed knowledge of news discourse (Bednarek & Caple, 2012). After all, news is primarily realised through words, which is why decisions concerning the language used in crisis communication have a great potential to influence people’s perceptions (Bednarek & Caple, 2012; see also Djerf-Pierre et al., 2014; Johansson & Odén, 2018; Mustafa-Awad & Kirner-Ludwig, 2017). Language is, therefore, a central resource when we seek to understand our social reality and communicate our experiences about health (Mahlberg & Brookes, 2021).

In extant research, communication about Covid-19 has been studied from different perspectives, all emphasising the crucial role of crisis communication. For example, communication and media scholars have investigated politicisation and polarisation (Hart et al., 2020), the effects of the pandemic on journalism (Quandt & Wahl-Jorgensen, 2021), communication dynamics (Lewis et al., 2021), public health communication (Nan & Thompson, 2021), news use (Ytre-Arne & Moe, 2021), and crisis management (Johansson et al., 2023). Furthermore, linguists have used multilingual and translational perspectives (Blumczynski & Wilson, 2022) and multimodal discourse analytical approaches (Tan & E, 2022). In their study, Ghersetti and colleagues (2023) found that news reporting in Sweden and Iceland mainly focused on health-related information from authorities during the first year of the pandemic. Their findings also indicated that criticism towards the presented information was not allocated much space in the news coverage. Attitudes and emotions in crisis communication have also been investigated. Hase and Engelke (2022) discovered that the news coverage in the UK in 2020 used fear-inducing discourse emphasising threats related to the pandemic. The discourse was used as a warning rather than a reflection of the ongoing situation. Meanwhile, the studies in the special issue edited by Mahlberg and Brookes (2021) concentrated on how attitudes and beliefs are built and mediated through language use in different contexts. For instance, Dong and colleagues (2021) discovered how the news media in various English-speaking countries typically used appreciation-oriented and emotional expressions when addressing the pandemic, which also correlates with the news media’s goal of attracting readers. Moreover, research has also concentrated on the reader’s perspective of the emotional responses the news evoked during the first lockdown (see, e.g., Eisele et al., 2022).

Notably, despite previous research, we still have limited knowledge of how the pandemic was represented at the news headline level. Given the vital role of news media in crisis communication and the power of news headlines to influence people’s perceptions (Eisele et al., 2022; Mustafa-Awad & Kirner-Ludwig, 2017; Quandt & Wahl-Jorgensen, 2022), we argue that it is necessary to understand what kind of pandemic images were created in the headlines. Therefore, we study the attitudes, emotions, themes, and actors in news headlines and aim to shed light on the evaluative images they create. In doing so, we build on existing scholarship on crisis communication by offering new knowledge on how this global health crisis was covered in Finnish news headlines. The extensive analysis of evaluation can also illuminate how much space attitudes and emotions gained in the news coverage. In this article, we address the following questions:

RQ1. How are different themes related to the Covid-19 pandemic evaluated in Finnish news headlines?

Typical of discourse analysis, we conducted a detailed qualitative examination and close reading of the news headlines that guide us in identifying evaluations, conceived of as evaluative parameters (Bednarek, 2010), themes, and actors. The data consist of digital news headlines in Finnish addressing Covid-19 in March 2020. The article is structured as follows: First, we compose our theoretical framework with the concepts of evaluation and crisis communication, followed by a description of the materials and methods of the study. Subsequently, we present our analysis by describing the evaluative parameters, themes, and actors identified from the headlines. Afterwards, we present three evaluative images of the pandemic that emerged at their intersection. Finally, we summarise the research in the discussion.

Evaluation is an inseparable part of language use. It serves as an umbrella term for the expression of a language user’s subjectivity, which can include attitudes, opinions, and emotions towards everything that can be evaluated through language (Bednarek & Caple, 2012; Thompson & Hunston, 2000). In the context of our research, evaluation not only reveals how the news writers and editors of headlines position themselves towards the reported matters but also reflects communal values (Thompson & Hunston, 2000). In other words, “identifying what the writer thinks reveals the ideology of the society that has produced the text” (Thompson & Hunston, 2000: 6). Evaluative language can be analysed through different frameworks, such as evaluation (Bednarek, 2006, 2010; Thompson & Hunston, 2000), subjectivity (Benveniste, 1966), positioning (Billig, 1987), stance (Du Bois, 2007; Jaffe, 2009), or appraisal (Martin & White, 2005). The frameworks seek to understand how language is employed by the actual speakers (Englebretson, 2007), but they have distinct analytical viewpoints that emphasise different aspects of language use (Bednarek, 2006). In this article, we apply the notion of evaluation and scrutinise the language of news headlines through evaluative parameters (Bednarek, 2010). Giaxoglou and Johansson (2020) have emphasised that the evaluative approach provides an effective tool for determining how certain interpretations appear and are presented, extended, and rooted into less stable attitudes and identities. Evaluative parameters are part of a framework specifically created to fit the analysis of news discourse (Bednarek, 2006, 2010; see also Thompson & Hunston, 2000). The parameters “refer to the standards, norms and values according to which we evaluate something through language” (Bednarek & Caple, 2012: 139), and they combine the previous frameworks mentioned above to become more suitable to news discourse, seeking to include a reasonable number of parameters in the analysis (Bednarek, 2006). For instance, the dimensions of the systemic functional linguistics of affect, judgement, and appreciation have been placed under one collective parameter of emotivity in the framework we utilise (Bednarek, 2006).

Communication during crises has been addressed widely in the social sciences and in communication and media studies, where the concepts vary from crisis and disaster communication (Palen & Hughes, 2018; Thompson & Hunston, 2000) to risk communication (Glik, 2007) and crisis reporting (Andén-Papadopoulos & Pantti, 2013). Crisis communication often deals with the impact crises cause on an organisational level (An & Gower, 2009; Coombs, 2007; see also Johansson et al., 2023); however, it can equally take a more comprehensive perspective. In this study, we follow Ghersetti and colleagues (2023), Hase and Engelke (2022), and Ytre-Arne and Moe (2023), who employed the notion of crisis communication to address news reporting during the Covid-19 pandemic. Expanding the concept beyond the organisational level allows us to recognise the news media’s crucial informative role in communicating with the public at different phases of a widespread crisis (An & Gower, 2009; Coombs, 2007; Ghersetti et al., 2023; Johansson & Odén, 2018).

In crisis communication, evaluation can be perceived as a strategy to construct appealing and effective news headlines (see, e.g., Molek-Kozakowska, 2013). During crises, information must be presented in a consistent yet attractive manner to guide citizens (Boin et al., 2016), for example, to follow recommendations. This also aligns with the discursive framework, which considers that headlines have a semantic and pragmatic double function (Iarovici & Amel, 1989). In other words, headlines are simultaneously summaries and advertisements, often presenting information in a bold, exaggerated, and polarised way (Andrew, 2013; Kronrod & Engel, 2001). The competition between newsrooms for readers and the increasing popularity of evening tabloids have also led to quality journalism and daily newspapers adopting characteristics of sensational reporting, previously exclusive to evening tabloids (Kantola, 2011a, 2011b; Rowe, 2010). Journalists and news writers analyse causes and consequences of newsworthy events, and they can emphasise entertainment, debate, storytelling, and emotions in their reporting (Deuze, 2007; Djerf-Pierre et al., 2014; Jaakkola et al., 2015; Johansson & Odén, 2018). For example, in the communication of the Covid-19 pandemic, appeals to emotions such as fear were entrenched (Hase & Engelke, 2022). The language of headlines also influences the perceptions of readers. Previous studies have illustrated that positive headlines can result in more favourable attitudes towards the reported events (Overgaard, 2021) than their negative counterparts (Wanta et al., 2004). Although the news discourse increasingly benefits from entertainment and emotions, communication by the news media still has a central position in crises, as people rely on news to provide information on the authorities’ decisions (Monahan & Ettinger, 2018; Prieto-Ramos et al., 2020). Ytre-Arne and Moe (2023) indicated how the news media remained a highly valued and trusted source of pandemic-related information during Covid-19, regardless of the complex information environment that citizens had to navigate. In the context of this study, we consider that news headlines play a key role in constructing effective crisis communication, especially in today’s digital era, where citizens tend to scroll through newsfeeds without necessarily engaging with the actual articles (see, e.g., Andrew, 2013). This role cannot be underestimated, because the news media have the power to shape the perceptions, and even the actions, of their readers (Djerf-Pierre et al., 2014; Johansson & Odén, 2018; Mustafa-Awad & Kirner-Ludwig, 2017). Fairclough (1995) treated news discourse as a means for social change.

News is not only produced by journalists and news editors, but there are also other actors who actively take part in the process. Broersma and colleagues (2013) describe that news is a result of a display of courtship between journalists and their sources. The news writer has to lure the source to contribute to the news story, give information, and provide interesting citations. However, the collaboration of different actors is often reciprocal: Journalists mediate stories while sources get visibility for their cause (Broersma et al., 2013; Johansson & Odén, 2018; Larsson, 2002). In the early stages of a crisis, news media collaborate with authorities, experts, and decision-makers as they all seek to inform the public about the current situation and the appropriate ways to act (Ghersetti et al., 2023; Johansson & Odén, 2018). These public figures are the most important source of information in a sudden crisis, and the media functions as the most effective way of passing the information (Eisele et al., 2022; Vultee & Wilkins, 2012). In addition, nongovernmental organisations and entrepreneurs can emerge in crisis communication, but for the latter, communication tends to concern the reputation of the company (Broersma et al., 2013). News coverage is often characterised by human interest stories, where ordinary citizens such as patients are given a space to talk about their personal experiences (Thorbjørnsrud & Ytreberg, 2020). These stories are typically emotional, evoke sympathy among readers, and ensure media attention, which further influences how people perceive reported events (Thorbjørnsrud & Ytreberg, 2020; Zillmann, 2002). Ordinary citizens are seen as authentic experience specialists who “speak from the real world”, in contrast to politicians, for example, who advocate vested interests (Thorbjørnsrud & Ytreberg, 2020). However, the authority figures remain visible actors in news because they add credibility to the communication in times of uncertainty (Boyce, 2006; Strömbäck & Nord, 2006). In the following section, we describe the data and methods of the study.

To study the evaluations, themes, and actors in news headlines and the evaluative images they created of the Covid-19 pandemic, we analysed the digital news headlines in Finnish from the digital news sources of Ilta-Sanomat, Helsingin Sanomat, and Yleisradio (Yle). Ilta-Sanomat is an evening tabloid, whereas Helsingin Sanomat is a local newspaper of the capital region of Finland. Both newspapers are published by Sanoma Media Finland, and they obtain revenues from advertisements and subscriptions. In contrast, Yle is a non-commercial public broadcasting company owned by the state. Yle has national television and radio channels and a website that contains news and more theme-specific sections. We selected the news sources based on a number of criteria. First, we wanted to add variability to the data and avoid concentrating on a specific region. For example, Helsingin Sanomat is a local newspaper and some of the news focus on the Helsinki region. Helsingin Sanomat was still included in the data because it also covers other news in a national and international level, and it is widely read in Finland. Second, the three sources allowed us to examine both commercial and non-commercial news media. Third, we made the decision regarding the news media based on the number of readers. The digital news platform of Ilta-Sanomat reaches over 2.9 million readers, Helsingin Sanomat is accessed by almost 1.8 million readers, and Yle has over 2.4 million readers per week (FIAM, 2023). Finally, we considered their relevance in the Finnish news media scene during the pandemic: All the news sources were at the forefront of the crisis communication as they provided information on the latest turns of events.

The news corpus was compiled by the TurkuNLP team at the University of Turku and consisted of billions of automatically collected words, including digital news, their headlines, and comment sections from 2018 until today. However, for our study, we only investigated news headlines from 16–17 March 2020. This specific timeframe was selected because it includes the historical declaration of the state of emergency. A similar decision was executed in Finland several decades ago and it now limited the basic rights of citizens, such as travelling inside national borders. As we do not seek to compare time periods or stages of the pandemic, we considered that a careful limitation of the data would allow us to closely examine how the crisis is represented in news media. Moreover, we wanted to obtain actual crisis communication for the analysis before any extensive emergence of political discourse and polarisation in the public discussion (see e.g., Ghersetti et al., 2023; Hart et al., 2020).

The compilation of the corpus started with the use of RSS feeds (continuously updated online feeds that allow users to access online news in a computer-readable format) from Finnish and Swedish news media. Many news media provide RSS feeds, including the news headline and link to the actual news article, on their website. The feeds also contain metadata in HTML, such as language used or photos that may not serve the research in question. Therefore, the Python package Trafilatura, designed for text extraction, was used to extract the necessary text (e.g., dates and news headlines) from the noisy HTML code for the purposes of our study. The Finnish data used in this research were extracted from a larger dataset using search terms such as Covid-19, korona [corona], and koronapandemia [Covid-19 pandemic]. The next step was to remove duplicates to increase diversity in the analysis. We started the analysis with 687 headlines; however, in the early stage, the data began to repeat. Therefore, the final data included in this study consisted of a random sample of 138 news headlines (1,606 words) in total. The headlines were divided as follows: 62 headlines from Ilta-Sanomat, 47 from Helsingin Sanomat, and 29 from Yle. The small size of the dataset does not seek to function in a representative manner but to offer an illustrative glance at the news headlines of the pandemic.

Our analysis was guided by a detailed qualitative examination and close reading of the data, typical of discourse analytical approaches. In other words, our analysis consisted of a sample of headlines that were manually studied systematically to find meanings within the text (Brummett, 2019). The process of close reading assisted us in the identification of the evaluative parameters (Bednarek, 2010) and themes. Throughout the analysis, we also recognised and marked different actors who had a voice in the headlines. We were especially interested in actors represented in the headlines other than news professionals. We consider these professionals as an inherent part of headlines, whether in the background with a neutral tone or more visible with an evaluative voice. Finally, we identified three evaluative images that emerged in the headlines at the intersection of the evaluative parameters, themes, and actors. To fit today’s news media discourse, the evaluative parameters synthesise previous frameworks of appraisal, evaluation, and stance (e.g., Martin & White, 2005; Thompson & Hunston, 2000). The framework we utilise contains twelve parameters for analysing evaluative meanings in news discourse. The parameters can also include subcategories that help specify the evaluation. For example, the parameter of emotivity has two subcategories from negative to positive emotivity, and the parameter of importance has subcategories from important to unimportant. Table 1 describes the most important parameters for our research (see Appendix A for all the parameters).

The framework has two parameters of sourcing – evidentiality and style – both of which are linked to the news source. More specifically, evidentiality’s subcategory of speech is related to the parameter of style, which means style provides information on how something was said by the source (Bednarek, 2010). Henceforth, for clarity, we refer to the parameters as evidentiality and style instead of the identical term of sourcing. According to Bednarek (2010), there are also similarities between other parameters: Emotivity and mental state are closely attached to emotions, but the main difference is that emotivity concerns a wider evaluation of different entities, whereas mental state is related to the mental states people can experience. Moreover, the parameters tend to co-occur, meaning one headline, expression, or word can contain several evaluative parameters. For example, an extremely negative evaluation of someone’s emotional mental state can also evoke negative emotivity. Evaluative meanings can be interpreted differently depending on the reader’s personal attitudes and experiences towards the discussed issues (Bednarek, 2008). We acknowledge the possibility of a certain level of subjectivity in the analysis due to the fuzzy nature of emotional language (Bednarek, 2008) and manual analysis. On the other hand, our analysis is based on decisions that have been consistent throughout the process, and we have executed several rounds of analysis to revise and strengthen these decisions. This has been possible due to the accessible amount of data. In the next section, we present our analysis based on the methods described above.

The importance of this study lies in the divergent practices of news headlines (Boin et al., 2016) and their power to influence how people understand and feel about major societal issues (Bednarek & Caple, 2012; Mustafa-Awad & Kirner-Ludwig, 2017). The systematic analysis of the evaluative parameters helps us better comprehend the attitudes and values in news headlines and helps us access them precisely through the three evaluative images. Primarily, our study suggests how the evaluations attached to different themes are reflected as evaluative images that can potentially influence how citizens perceive the Covid-19 crisis. First, the pandemic is represented as a crisis that evokes concern and solidarity in the everyday lives of citizens. Second, the pandemic is described as a challenging health crisis with authorities as responsible decision-makers. Third, the pandemic is portrayed as a crisis that creates concern and has negative impacts on the different functions of society.

As for the actors involved in crisis communication of news media, it is worth noticing that news headlines are always produced by news professionals, represented as objective reporters or evaluative voices. In addition to them, actors in our data include parents, family members, and people from risk groups. The findings support Thorbjørnsrud and Ytreberg (2020), who noted that ordinary citizens’ personal experiences of health are important in the news media because they attract readers and arouse sympathy. Further, healthcare actors, politicians, other decision-makers, researchers, and company managers are represented in our headlines. Boyce (2006) stated that the appearance of authorities and different experts in news media is crucial because they offer fact-based information and enhance credibility. Additionally, we propose that the presence of authorities offered reassurance to citizens under new and uncertain circumstances. In fact, the citizens’ trust towards authorities, institutions, and news media in Finland during the first year of the pandemic was relatively high (Jallinoja & Väliverronen, 2021), which would support the use of authorities and experts in news media coverage.

Our findings support extant research showing that during crises, news media regard its role as having dual responsibilities of distributing relevant information from the authorities and experts to the public (Eisele et al., 2022; Ghersetti et al., 2023; Vultee & Wilkins, 2012) and deploying evaluative language on reported events (Boin et al., 2016; Eisele et al., 2022; Ghersetti et al., 2023; Overgaard, 2021; Wanta et al., 2004). In contrast to Ghersetti and colleagues (2023), our findings show how news reporting in Finland covered themes beyond health. Furthermore, the results of this study highlight the important role of individual headlines as mediators of crisis information, stories, and experiences. That is, the results reflect the power of compact headlines within the wider news genre. The evaluative images presented in this study demonstrate how practical aspects of the pandemic, such as decision-making and information that seeks to guide citizens, co-occur with the emotional and personal side of the citizens’ lives at a grassroots level (see also Ghersetti et al., 2023; Hase & Engelke, 2022; Lee, 2022). In other words, the public and private spheres overlap in the headlines. Our findings contribute to clarifying the complex media environment during a crisis, where different types of information are presented (see, e.g., Ytre-Arne & Moe, 2023) through emotional and attitude-oriented expressions.

Our results could serve actors ranging from decision-makers and experts to ordinary citizens for future purposes. The research may give preliminary ideas to journalists and newsrooms about the evaluative output of their crisis response, and whether the significant role of attitudes and emotions in the headlines appear desirable or not. The results could also be used to reflect the role and visibility of different actors in crisis communication and could also open discussions between newsrooms and public authorities on whether there were aspects of the pandemic that should have been discussed more comprehensively or vice versa. The outcomes of these considerations could be adapted and realised in health crisis communication of the future. Finally, the study may be of interest to ordinary citizens who were represented in the headlines and who consume the news.

With this article, we have demonstrated how the global health crisis was covered in the Finnish news headlines. Our study builds on existing scholarship on health crisis communication in news media and sheds insights through a linguistic lens into the role of attitudes and emotions in the news headlines communicating a new crisis. The question remains whether citizens perceived the pandemic as a multi-effected health crisis, as described in the headlines. This would open possibilities for future research to delve into the relationship between news media representations and citizens’ personal perceptions, and whether they correlate or differ. In addition, it would be beneficial to examine news headlines during a longer time period to detect changes in attitudes and emotions, especially if the news media’s criticism towards various restrictive measures emerged in the later stages of the pandemic (Ghersetti et al., 2023). Moreover, conflicting opinions, disinformation, and conspiracy theories attached to the pandemic over the past years would offer fascinating and essential avenues for future studies to deepen our knowledge of health crisis communication.