Pandemic information flows

Co-authored by the Brexit and Belonging team, and co-published across Media Effects Research and Brexit and Belonging.

In times of crisis when responsible action is required, we need clear and credible information. Our analysis shows that coronavirus news is produced in a concentrated network where news and information providers as well as the British public heavily rely on the BBC as a primary source.

With a number of running data initiatives at our disposal, including large-scale media mapping, passive metering of news browsing, and ethnographic fieldwork, we turned to explore some of the difficult questions related to the media coverage of Covid-19 in the UK and in particular how citizens manage the overabundance of information that is characteristic of the current “infodemic”.

Since the outbreak of the pandemic, the volume of stories mentioning coronavirus-related keywords* have grown exponentially, with the biggest spike between 8–19 March. By the end of March, three quarters of the entire online news landscape in the UK contained at least some mention of the virus.

Many fear that this abundance of information is anything but helpful as the public tries to navigate through an existential crisis, one that in the UK swiftly came after a news saturated Brexit crisis. Misinformation has been a pressing concern in both instances, with current coronavirus issues including the misrepresentation of preliminary scientific evidence, incorrect advice on causes, symptoms, and treatment.

To understand the scale of the problem, we turn to a coronavirus corpus of 215,952 news stories**. This contains some aggregate data on information flows we thus explore (1) the number of shares on Facebook aggregated per domain, and (2) the hyperlinks extracted from these stories. Then complementing this, we revisit (3) ethnographic fieldwork data from across the South West, North East and Midlands areas of the UK to gain additional insights into how citizens interact with, and make sense of, vast amounts of media coverage in practice.

In addition to this, we will also investigate the textual content of our stories to track coronavirus misinformation more specifically. We aim to blog the first results in the coming weeks.

(1) As the coronavirus pandemic unfolds, the British public increasingly turns to the BBC for information

​Traditional sources of news such as broadcast news (BBC, Sky), and the more trusted partisan broadsheets such as the Telegraph and the Guardian are not able to keep up with the excess volume of news stories (see Y axis) generated by their online competitors (Yahoo), including the tabloid press (The Sun and Express). The volume of news produced, however, does not indicate whether the public finds it credible and of value. We find that over time, people were much more likely to share the stories, though small in volume, produced by the BBC. Social media sharing is an indication that the source is trusted and by this measure the BBC is far and above the most trusted source for Covid-19 information.

In that critical period of exponential growth in story volume in March 2020, the BBC emerged as the primary source of information. Our animation above shows this dynamic clearly as the BBC gradually shifts away from the rest of the sources around mid-March. By the end of the month, the BBC counted nearly three times as many shares on Facebook than its closest competitor out of 4,397 sources. These sources may be explored in detail by zooming into and hovering over the markers in the animation above.

The BBC’s strategic position has been noted by the UK government. PM Boris Johnson started his term in office with a boycott of BBC Radio 4’s Today Programme due to concerns about its impartiality. In addition, he accelerated attempts to shrink one of the most venerated news organisations around the globe by threatening to cut its primary source of funding by removing the license fee altogether.

The boycott has, however, ended during the pandemic and we would argue this was a wise and strategic decision. The BBC represents an opportunity for a shared news experience by the public. ​We think the public uses the BBC’s unmatched source credibility to navigate in times of crisis (and elections). Importantly the BBC can counteract any polarisation in news viewing experiences that is generated by social media.

Our findings resonate with a recent Ofcom survey reporting 82% of adults online using the BBC to find information about Covid-19.

(2) News coverage of Covid-19 represents a concentrated news production network led by the BBC

We extracted all hyperlinks found in this corpus of news stories to draw a directed network representing the sources used to report on the coronavirus. For example, a news story from the Telegraph that contains a link to a BBC article is represented by the two “nodes” (BBC and Telegraph) and the linking edge: Telegraph ➡ BBC where the direction of the arrow indicates that the Telegraph is relying on the BBC as a source. We believe networks such as these offer a way of representing inter-media news production and agenda setting. In our example, the production of news in the Telegraph builds on the BBC. In our corpus of data, we found 10,251 such links between news sources.

Nodes (domains) are sized based on their in-degrees i.e. the number of links pointing to them from other sources.

As we would expect, the government web pages are the main source of information for other news outlets. But among news organisations, the BBC represents a main source of information in the news production environment. Our key point is, then, that the BBC provides an important source of credible information for individuals and is shared widely. Additionally, it is an important source for other news and information providers.

The entire Covid-19 coverage seems to be drawing on one main source namely the UK government’s communications. There are a large number news outlets and websites that connect to this node directly (many of them local news domains). Domains that have a traditional print format as well, however, seem to connect to this node indirectly via the BBC (the vast majority of them) but also via the more trusted and shared sites we already plotted above.

(3) A view from ethnographic research on Brexit and everyday media practices

The qualitative data comprise fieldwork notes from an ongoing ethnographic project as well as 180 interviews conducted in the South West, North East and Midlands areas of England. The fieldwork began in October 2018 and ended on January 31st 2020. The project explores questions of identity, belonging, and importantly for this blog, the role of media in “Brexit Britain”. Our focus is on the diverse types of knowledge that individuals mobilise to engage with media representations of Brexit. This process is contextualised within participant observation of individuals’ daily media practices with a focus on practices relating to news on Brexit and its outcomes.

One key theme emerging from our fieldwork across England is that the BBC was seen as a problematic source of information by both leavers and remainers. In this regard, people on both sides of the Brexit debate argued that the same BBC programmes were politically biased. For example, BBC1’s Question Time and Radio 4’s the Today Programme were thought by both leavers and remainers to be biased towards the side they did not support. Question Time was accused of having too many leavers or remainers on the panel and had was described as having ‘planted’ too many leavers or remainers in the audience. This marked mood of distrust of the BBC during the Brexit crisis renders the current return to the BBC as the most trusted source during this time of lockdown very significant.

In addition, it is worth highlighting that while people might be reading, sharing and discussing information on the internet during the lockdown, this does not necessarily mean they believe it whole heartedly. A key example here is that some ardent leavers we have spoken to did not straightforwardly believe the Brexit bus slogan “£350 million to the NHS”, rather they saw it as symbolic of the money that they thought the British government wasted on the EU.

It is also important to note that while it might appear that people are turning to the BBC during the pandemic, this statistic does not account for the ways in which people might also be reflexively managing their media consumption during this period of lockdown because they find the media emotionally distressing. For instance, we found that some people we spoke to about Brexit actively avoided media coverage on Brexit because the representation of the nation, national politics, disunity and governance was too upsetting. Likewise, similar (and hidden) patterns might well be at work now over the coronavirus coverage.

And finally there is something insightful to say from an ethnographic point of view about the idea that the social sharing of information is an indication of trust and source credibility. We found in our research on Brexit that people used social media e.g. Twitter, WhatsApp and Facebook to share news and views on Brexit. Facebook exchanges could at times cause tension and falling out amongst neighbours that were reproduced in everyday neighbourhood interactions. At the same time Facebook also united people who might not otherwise have known each other in communities of belonging and support around Brexit and this was the case especially for activists.

It would seem for Covid-19 that social media is being used as a way to connect people that might not have connected before via this medium because they are now having to socially distance and isolate from each other Facebook and other forms of social media are also being mobilised to share news, help and information that is locally relevant e.g. neighbourhood Facebook groups share news about which shops are offering home deliveries and how to volunteer in the community. Some forms of social media are also being used as sites to signal that help and care from those nearby is needed, and building new networks of social reliance and connection.

We aim to submit the underlying working paper to SocArXiv within the next few weeks. Please check back later for updates. In the meantime, feel free to cite this blog as source or contact us for more information.

* “coronavirus”, “covid”, “ncov”
** a near-complete coverage, but important exceptions apply. The Daily Mail has requested that Webhose API, our data source, do not crawl their content in our period of analysis. So far we know of no other major source missing.