Media portrayal of panic buying: A content analysis of online news portals

All the search engines have been identified by the investigators. Three independent investigators searched the allotted search engines using the “panic buying” keyword. We searched, collected, and analysed the news reports published in English discussing the PB. The search was done between 23 and 30 May 2020. All online news articles, channel interviews, news forums, personal views, and panel discussions (online published non-video reports) were included. Journal articles, journal posts, and social media posts were excluded. After the data collection was completed, two independent researchers carried out cross-checks of the data and data cleaning. Duplications were checked and removed by tracing the date of publication and title of the reports. Also, we removed the responses that were analysed in the initial phase study (Arafat et al., 2020a).

We used the same instrument that was developed during our previous study (Arafat et al., 2020a). The instrument included country of publication of the report, country to which the news report relates, type of newspaper, newspaper name, area (city) of publication, dates of publication, the scarce object for panic purchase, media report material, and relation to similar events of panic purchase in the past. The news that reported no specific countries, multiple countries, or a mix between developing and developed countries was excluded from the analysis. We evaluated the positive and negative aspects of media reporting based on the contents available in the reports. The list of positive and negative aspects was finalized prior to the beginning of data analysis by the process of consensus building between the authors (qualified psychiatrists/public health experts with clinical and research experience of working with PB) based on the face validity of selected queries.

We compared the reporting patterns in the countries divided on the basis of the Human Development Index report by the UNDP. In terms of percentages and proportions, the data collected were presented. A chi-square test was used to make distinctions between the LMICs and high-income countries (HICs). The p-value of <0.05 was found important in this analysis.

From 23 May until 30 May 2020, a total of 525 news reports regarding PB were analysed. The vast majority of the news came from the US (22.1%), followed by the UK (18.5%), India (15%), and Australia (5.5%). Each of the other countries has in total less than 10 reports. Approximately half of the news (48%) specified the scarce object in the report. The scarce objects include essential medicines, masks, hand sanitizers, gloves, toilet paper, groceries, fuels, food grains, and even firearms and ammunition. Concerning the context of PB news, the vast majority were related to the recent COVID-19 pandemic (95%) and the remaining include disasters and economic events (1%) and other events such as festivals, fuel price hike, and even war (<1%).

Concerning the positive aspects of news content posts related to PB, approximately three-fourth of the news mention the impact of the PB (71%) and approximately half (49.3%) discuss the government action to handle the PB issues. Furthermore, around one-third (36.4%) discuss the expert opinion regarding PB and approximately one-fifth discuss the psychology of PB (20.6%), the rumour about PB (21.5%), and suggest the remedial measures (18.5%).

Concerning the negative aspects of the news content, the vast majority of post titles refer to PB (96.6%) and approximately three-fourth of the news mention the cause (75.4%). Pictures or images of people in line for purchasing stuff or empty groceries were also common (62.3%). The public's PB behaviour was blamed by around a quarter of the news posts (26.3%). People involved in business also became the object of blame by some media (9.3%). Meanwhile, a few media blamed the government for the PB action (8%). The details of the content analysis are summarized in Table 1.

Media in LMICs were 1.5 times more likely to include expert opinion with regard to ongoing PB (x² = 4.7, p = 0.03) and 2.1 times more likely to discuss the rumour regarding PB (x² = 11.4, p = 0.001) as compared to HICs. However, the media in LMICs were almost thrice more likely to report the cause of PB (x² = 14.3, p = 0.001) and also thrice more likely to mention the impact of PB (x² = 19.6, p = 0.001) as compared to HICs. Meanwhile, the media in HICs and very HICs were almost twice more likely to report similar incidents in the past or elsewhere (x² = 7.63, p = 0.005). The details of the crude odds ratio of each content of news reports and their association are presented in Table 2.

Media reporting has an influential role in PB from spreading the behaviour to controlling it (Arafat et al., 2020b). There are no comparable reports analysing media coverage, and there is also a scarcity of evidence exploring the media portrayal of PB.

To the authors’ best knowledge, this is the first study to report the media portrayal of PB. The paper also assessed the variations of media reporting on the basis of income of countries. Although a large number of expert opinions and media reports have been published as non-scientific literature, this study is among the handful of scientific studies that have looked into an important area of media reporting of PB.

The current study has several limitations. First, we only included the reports published in English. The reports in non-English languages have been completely excluded, which represent a significant chunk of media reports. Second, there might be biases regarding the country reporting the PB as the search engines tend to pick up the news that is read widely. The PB present in small countries with limited Internet access and less powerful websites might have been missed in this study. The search for reports is based on the search engine; the more popular the website for reporting, the more the chance to be picked up. Third, the psychometric property of the instrument was not tested.

Certainly, empirical studies exploring its several aspects are warranted to understand the phenomenon as well as to consider its negative effects. Also, further studies in this area from the consumer perspective are needed as well (Habel et al., 2020). There is a need for reporting guidelines for the PB as it is a mass behaviour and occurs mostly at the time of public health or other emergencies. As there is no guideline that addresses the media reporting on PB, we may suggest several recommendations such as that media should not sensationalize the mass panic behaviours like PB, catastrophize, and post the photos and videos of scarce objects and empty shelves. Side by side, media should avoid blaming, address the worries, anger, and frustration of mass people, avoid mentioning the public reaction in news reports, and avoid interrogating people on sensitive matters related to PB.

Certain aspects of PB may be the potential source of challenges such as that the behaviour is usually sudden, happens during emergency events, the items could vary, the behaviour could be rational, and the problem is under-researched (Arafat et al., 2020d; Tsao et al., 2019; Kulemeka, 2010). Moreover, it is fundamentally difficult to predict the event (Arafat et al., 2020d; Tsao et al., 2019; Kulemeka, 2010). However, mentioning the negative characteristics in the reports would increase fear and insecurity, which in turn increases PB and destroys the equilibrium. At the same time, reporting the helpful characteristics could reduce fear and insecurity even though there is a shortage. For example, if the media mentions that the government is trying to maintain the supply, then fear may be reduced.

Panic buying is a dynamic phenomenon. As the COVID-19 pandemic progresses, the needs of people change and disaster preparedness also changes. The media also lose interest in PB, and the focus moves to other important things as the pandemic progresses. There is a need for the sequential evaluation of media reports on PB to gain better insight into this topic.