In 2020 the coronavirus pandemic has dominated the headlines and has received an incredible amount of media attention. Many different media outlets such as newspapers and social media channels have been publishing daily updates about the virus, concerning its development and its spread, the number of infections and the number of deaths in each country, and its consequences. Research by Mertens, G., Gerritsen, L., Duijndam S., & Salemink, L. (2020) and Garfin, D. R., Silver, R. C., & Holman, E. A. (2020) suggests that the media coverage about the global pandemic could potentially predict fear and anxiety among people. In fact, research by Asmundson, G. J. G., & Taylor, S. (2020) suggests that fear among populations increased during the coronavirus crisis. For this reason, we, the researchers, considered it interesting to conduct an analysis comparing how fear about the coronavirus, also known as Covid-19 has been portrayed in two different newspapers. The research question for this content analysis is: “Is there a difference in the way that The Washington Post and The Daily Telegraph (London) portray fear about Covid-19 from the 1stof January of 2020 till 31stof May?”. This research question aims to determine whether there is a significant difference between the way the two different newspapers frame fear in their news articles. Moreover, it focuses on the comparison of an American newspaper, The Washington Post, and a British newspaper, The Daily Telegraph (London). This is to examine whether the analysis also shows a contrast between the two English speaking countries. The main objective of this study is to analyse how fear about the coronavirus is framed in the two newspapers, and if there are diverse threats present that encourage the frame of fear.
This is interesting to analyse because it could be beneficial for society to understand how information and news can affect people’s behaviour and their emotional well-being. In today’s society, the amount of information that can be found online is incredible, in fact, information overload can be quite problematic for people. This is because it can give uncertainty about the entire issue since it is difficult to determine which source to trust. Additionally, this study could also help raise awareness about the different types of threats as well as the fears among different people due to Covid-19. For example, the rise of fear related to the uncertainty about the virus since it is completely new, or its economic consequences for individuals as well as society as a whole. These factors could also enable people to notice the important role that the media play in society, as well as provide a deeper understanding of how public opinion is formed.
Apart from this, this content analysis could contribute to the current literature about the coronavirus crisis, since it remains a novel topic. As mentioned earlier, current research shows that media exposure could increase different types of fear among individuals. However, there are not many studies dedicated to understanding the role of different news frames and how it could potentially shape people’s perceptions about the coronavirus. This is why this study could provide more information about the different types of news frames used to portray fear about coronavirus in society. Furthermore, it could also contribute to the current research about the differences between the news frames used in American and British newspapers.
The number of infections and deaths in the United States (US) and in the United Kingdom (UK) could also have had an influence on the reporting of the coronavirus in the two countries. Perhaps the fact that the US suffered more from the virus could mean that there are more news frames present about the possible threats and fears from the virus. For this reason, our hypothesis is as follows: “The Washington Post includes more articles with frames portraying fear about the coronavirus in comparison to The Daily Telegraph.”
METHOD
2.1 Measurement of concepts and inter-coder reliability
The aim of this research is to identify how the Washington Post and the Daily Telegraph portray fear in their articles about Covid-19. The variable fear and its relevant indicators, ‘economic threat’, ‘health threat’, ‘crime threat’, and ‘other threat,’ are applied to this study from the threat frame used in the study by Mertens et al. (2020). ‘Economic threat’ indicator includes fear of the decline of the economy and the uncertainty of economic stability. It entails the worry of job loss and economic uncertainty concerning companies, such as companies shutting down or losing profit. Furthermore, the ‘health threat’ indicator describes the fear of being infected with Covid-19, receiving insufficient medical treatment, and being diagnosed with mental health issues. For the ‘crime threat’ indicator, anything that threatens someone’s safety is considered a crime, including racism, robbery, or violence, both physical or verbal. Lastly, indicator ‘other threat’ includes serious threats outside the range of previous indicators, such as the fear of death, food shortages, or receiving misinformation.
The coding scheme contains a dichotomous scale for every indicator, whereby number 1 (threat identified) was selected if any indication of the particular threat was present and number 0 (threat not identified) chosen only if there is absolutely no indication. Administrative variables are ‘number of coders’ (1=Laura, 2=Hope, 3=Natalia, and 4=Sofia) to identify the coder, and ‘article ID’, which receives a numerical code (for example, 743) that is based on the predetermined ID of an article. Additionally, ‘date of the article’ (dd.mm.yyyy) indicates the day the article has been coded and ‘title of the article’ is coded by copy-pasting the title directly from Nexis Uni.
In order to test for an inter-coder reliability in our content analysis, we coded about 10 articles as a group. By modifying our codebook together, we were able to change a few aspects of the codebook inspired by Mertens et al. (2020) and agreed on the final codebook used to code. After getting a more valid codebook, we conducted an inter-reliability test in which we tested for the Krippendorff’s Alpha coefficients on SPSS. The results we found were the following: αeconomic = .97 with a 1% chance of failure to achieve an alpha of at least .90; αhealth = .45 with a 66% probability of failure to achieve an alpha of at least .50 and a 100% chance of failure to achieve an alpha of .80 and above; αcrime = .88 with a 50% probability of failure to achieve an alpha of at least .90 and 3% chance of failure to achieve an alpha above .70; αother = .57 with a 100% probability of failure to reach an alpha of .90, and 20% chance of failure to achieve an alpha above .50. Hence, the Krippendorff’s Alpha is considered acceptable for ‘economic threat’ and ‘crime threat’, but not for ‘health threat’ and ‘other threat’. To conclude, our measuring device is reliable on certain aspects such as economics and crime, but not for others like health and other threats.
2.2 Search phrase optimisation procedure
Firstly, the search filter was set to provide information on the two targeted newspapers and our specific timeline. Next, it was important to include ‘(coronavirus or covid!)’ to catch all the articles that mention the virus. Mentioning ‘coronavirus’ alone would have been insufficient as it is the generic term of all viruses that have the specific crown-like halo, such as SARS (CDC, 2020). Nevertheless, ‘coronavirus’ was the used term for the virus at the beginning of the outbreak before it was named Covid-19, making it still a relevant search term to successfully retrieve meaningful articles.
The initial search phrase contained terms that could enhance fear in the articles: ‘(coronavirus OR covid!) AND (worry OR fear)’. It turned out that it was not adequate enough since newspapers use various other terms to highlight their messages. Therefore, the term ‘anxiety’ was added, which increased the number of articles. After browsing through the articles selected by ‘the Google random number generator’, the words ‘risk’ and ‘danger’ were found repeatedly in the articles since they were used to emphasise the threats that Covid-19 has on various areas of life. The detected words were added leading to the final search phrase: ‘(coronavirus OR covid!) AND (worry OR fear OR anxiety OR risk or danger)’, which resulted in 7746 articles that after randomly browsing through, provided the content that would be best suitable for this particular analysis.