As vaccines start rolling out, here’s what our research says about communication and coronavirus

Our research team leader on what journalists and governments should know as we enter the next phase of the pandemic
Dr. Tommy Wong gives a thumbs up as he receives one of the first vaccinations at Mt. Sinai Hospital from Pfizer-BioNTech during the coronavirus disease (COVID-19) pandemic in the Manhattan borough of New York City, New York, U.S., December 15, 2020. REUTERS/Carlo Allegri

Dr. Tommy Wong receives one of the first vaccinations at Mt. Sinai Hospital in New York City. REUTERS/Carlo Allegri

Dr Richard Fletcher

The coronavirus situation represents both a health crisis and a communications emergency – especially as we enter the vaccination phase. With a significant minority expressing a degree of hesitancy about receiving a coronavirus vaccine, clear and effective communication about the benefits and risks (or lack of) will undoubtedly be important for maximising uptake. At the same time, many familiar communication challenges – such as convincing people of the need to limit social contact, to wear face coverings, and to wash their hands – will be just as important for much of 2021 as they were in 2020. Perhaps more so, if the vaccine rollout is accompanied by a sense of complacency about taking basic precautions.

Here, I will describe some of the empirical findings from our UK COVID-19 News and Information project to highlight what we learned about communication and coronavirus in 2020, and what it means for the vaccination phase. As such, the focus is on audience research through nationally representative online surveys, and this is not a comprehensive record of all communications research on coronavirus that exists, but I will draw on other studies to expand the scope of the findings.

1. The news media is central to coronavirus communication

First, we know that the news media is absolutely central to how people get information about coronavirus. At the beginning of the pandemic in March 2020, our online survey of six countries (Argentina, Germany, Spain, South Korea, the UK, and the US) found that news organisations were the single most widely used source of information about coronavirus, ahead of governments, scientists and doctors, health organisations, and ordinary people (Nielsen, Fletcher, Newman, et al. 2020). When surveyed, between around 50% and 75% said they had accessed coronavirus news from news organisations, depending on the country, as Figure 1 shows. Furthermore, it is highly likely that information from scientists, governments and health organisations was in part accessed through coverage from the news media.

Figure 1

In the UK, ongoing research by communications regulator Ofcom has consistently found throughout 2020 that traditional broadcast and print news outlets (accessed either online or offline) are the most widely used sources of news about coronavirus – ahead of social media, messaging apps, and other online-only news sites. Ofcom also found that BBC services account for much of the coronavirus news use in the UK (Ofcom 2020).

Search engines, social media, messaging apps and video sites are used for coronavirus news by a significant minority. But as is the case with news use more generally (Newman et al. 2020), the combined online and offline coronavirus news reach of traditional outlets is often larger (Nielsen, Fletcher, Newman, et al. 2020). Social media news use is more widespread among the younger age groups, but most come across coronavirus news on social media ‘incidentally’ as they are using it for other purposes (Nielsen, Kalogeropoulos, and Fletcher 2020b).

2. Coronavirus news use is varied and unequal

Coronavirus news use surged at the beginning of the first wave in the UK in April, but decreased substantially over the summer. The proportion who got news about COVID-19 at least once a day per week on average dropped by 24 points from 79% in mid-April to 55% in mid-August (Nielsen, Fletcher, Kalogeropoulos, et al. 2020). It’s not clear whether this surge in news use was primarily driven by a need for reliable information about coronavirus, or by the fact that people had more time for certain activities during lockdown.

News avoidance also grew in the early stages of the pandemic in the UK. In mid-April 15% said that they always or often avoid the news these days, rising to 25% in mid-May and staying there over the summer, even as lockdown ended and the crisis in many ways became temporarily less acute (Fletcher, Kalogeropoulos, and Nielsen 2020a). Two-thirds of those that actively avoid the news say it is because it has a bad effect on their mood (Kalogeropoulos, Fletcher, and Nielsen 2020b).   

Both the decline in news use and the rise in news avoidance have created inequalities in coronavirus news use, as Figure 2 shows. The much higher levels of news use we saw at the beginning of the crisis in the UK partly hid longstanding inequalities in news use around age and gender, but these had re-emerged by June (Fletcher et al. 2020). In April there was a 12-point gap between the proportion of 55s and over (86%) that consumed coronavirus news at least once a day on average and those under 55 (74%), but this had widened to 24 points by the end of June (75% vs. 51%). The gap in coronavirus news use between men and women grew from 1 point to 8 points over the same period. Gaps by household income and education were real, but did not grow or shrink. 

Figure 2

Information inequalities in the UK around COVID-19

It is important to keep in mind that different demographic and political groups access coronavirus news in different ways – again, usually mirroring patterns we see in normal news use. Older people are more likely to use television and print. Younger people are more likely to use social media. The use of certain news outlets also follows patterns we see from research on selective exposure, whereby those on the right will be more likely to use right-leaning sources, and vice versa. What is clear is that no one news source or news outlet can reach everyone.

3. People trust health experts, but are more cautious about the news media

The traditional news media are the most widely used source of coronavirus news, but they are not the most trusted. At the start of the pandemic, news organisations were typically less trusted on coronavirus than scientists, doctors, and health organisations, but were ahead of politicians and ordinary people (Figure 3). News organisations were roughly as trusted as national governments, with between 50% and 70% scoring them 6 or over on a 0-10 trust scale (Nielsen, Fletcher, Newman, et al. 2020).

Figure 3 

These levels of trust are higher than we have come to expect for news more generally (Newman et al. 2020). This may reflect both the nature of the coverage and the seriousness of the crisis. There may also have been a brief ‘rally around the news’, similar to how support for some political leaders rose sharply in March (Jennings 2020).

In the UK, however, trust in news organisations as a source of news and information about coronavirus quickly started to fall, alongside trust in the government (Fletcher, Kalogeropoulos, and Nielsen 2020b). Trust in news organisations as a source of news and information about coronavirus fell from 57% in April to 46% in May, and stayed at similar levels over the summer. By August there were similar falls in the proportion that thought the news media helped them understand the pandemic, and helped explain what they should do in response. Around one-third (35%) thought that the coronavirus situation had been made worse by how the news media covered it (Nielsen, Kalogeropoulos, and Fletcher 2020a).

Although trust in news organisations as a source of news and information about coronavirus has waned, it is much higher than the equivalent trust in information found via social media, messaging apps, search engines, and video sites (Nielsen, Fletcher, Newman, et al. 2020). We find consistent “trust gaps” between trust in coronavirus information from news media and information from platforms, often about 20 points between search engines and news media, and about 30 points between social media, video sharing sites, or messaging applications and news media.

4. Misinformation problems are real and serious, but small part of overall media use and often driven by elites

Misinformation has been one of the most heavily researched aspects of the crisis. There are real and serious misinformation problems associated with the pandemic and these problems are especially visible on big platforms like Facebook and YouTube. Our analysis of the work of fact-checkers in the early months of 2020 showed a huge surge in coronavirus misinformation, much of it true information that was spun, twisted, recontextualised as opposed to completely fabricated (Brennen et al. 2020). Visual forms of misinformation tend to be relabelled as opposed to manipulated, and manipulation was done using simple tools rather than “deepfakes” (Brennen, Simon, and Nielsen 2020).

Most coronavirus fact-checks address misinformation that has come from ordinary people. However, ‘top-down’ misinformation from politicians, celebrities and public figures receive the most engagement on social media (Brennen et al. 2020). This is partly reflected in our data from the UK, where the proportion of those who are ‘very’ or ‘extremely concerned about misinformation from politicians and the government rose to around 40% in August, overtaking ordinary people in April (Nielsen, Kalogeropoulos, and Fletcher 2020a) (Figure 4). Elites play a key role in misinformation problems, and the public is keenly aware of this.

Figure 4

One of the most challenging tasks for researchers in this area is identifying how many people actually see false and misleading information about coronavirus. The simplest approach is to just ask people in a survey, and although this has obvious limitations, it is often the only realistic option.


In March we found that self-reported coronavirus misinformation exposure was low. Figures vary a lot by country and by source, but it is usually a minority of between 20% and 40% who say they have come across a lot or a great deal of false or misleading information about coronavirus. This is broadly in line with tracking studies from the US that tend to find low levels of misinformation exposure more generally (Nelson and Taneja 2018). A recent study from the US found that just 18.5% visit a vaccine-sceptical webpage each year (Guess et al. 2020). More broadly, even in the extreme case of the US, “fake news” is a very small part of most people’s media use (less than 1% of news use on average, and less than 0.2% of all media use) (Allen et al 2020).

People tend to identify governments, politicians, news organisations, and ordinary people as the main sources of misinformation, but few reported seeing misinformation from scientists, doctors and health organisations (Nielsen, Fletcher, Newman, et al. 2020). People also tend to associate misinformation exposure with social media, and to a lesser extent, video sites, messaging apps and search engines.

5. Knowledge and behaviour is partly shaped by news and information

One reason why misinformation matters in this case is the potential for negative effects on what people know about coronavirus, and ultimately their willingness to follow rules and take preventive steps (Hameleers, van der Meer, and Brosius 2020).

Before turning to this, it important to make clear that most people are quite knowledgeable about coronavirus (Nielsen, Fletcher, Newman, et al. 2020; Nielsen, Fletcher, Kalogeropoulos, et al. 2020), and belief in many frequently-discussed coronavirus myths – such as the link with 5G technology – is rare (Roozenbeek et al. 2020). Most people also say they follow the rules, though some feel confused about what they are supposed to do (Kyriakidou et al. 2020), and people may find them difficult to follow to the letter in practice.

Our own research has found that news use is typically associated with higher coronavirus knowledge (Nielsen, Fletcher, Newman, et al. 2020), following government guidance, and willingness to take preventive steps (Kalogeropoulos, Fletcher, and Nielsen 2020a), as Figure 5 shows. Other studies from the US have reached similar conclusions (Motta, Stecula, and Farhart 2020; Bridgman et al. 2020), but have also found different effects for different news outlets (Hall Jamieson and Albarracín 2020).

Figure 5

Research from the US has also found that social media use is associated with lower coronavirus knowledge (Pew Research Center 2020; Bridgman et al. 2020), but our own research from six countries (including the US) did not find significant patterns (Nielsen, Fletcher, Newman, et al. 2020).

Studies have found that trust in scientists and experts is associated with increased coronavirus risk perception (Dryhurst et al. 2020) and lower susceptibility to coronavirus misinformation (Roozenbeek et al. 2020). The effect of trust in the news media is less clear, but some studies find positive links (Loomba et al. 2020) and we should keep in mind that scientists and experts rely in part on the news media to communicate with the public, making it difficult to disentangle the effects.

More broadly, the positive effects of news use likely hinge on whether people trust it enough to properly take the information in. It is for this reason that we have identified the ‘infodemically vulnerable’ – a small but growing group that make little or no use of news about the pandemic, and have low trust in it even if it reaches them. In the UK, the size of this group grew from around 6% in April to 15% in August (Nielsen, Fletcher, Kalogeropoulos, et al. 2020).

6. Communication is essential to the vaccination phase

This is crucially important for the vaccination phase of the pandemic. Much research, including our own (Kalogeropoulos, Fletcher, and Nielsen 2020a), highlights a significant minority who are coronavirus ‘vaccine hesitant’ (Lazarus et al. 2020). But as Brendan Nyhan put it recently, many countries start from a position of strength because, although some express a degree of hesitancy, actual vaccination rates are high. And, as Samantha Vandersloth and her colleagues point out, while communications and public perception is important, “widespread public concern for vaccine safety does not appear to be strongly correlated with vaccination rates.”

Figure 6.

But nothing should be taken for granted. In addition to the continued importance of trust in scientific and medical expertise (Freeman et al. 2020), a recent experiment found that although trust in television news in the UK and the US was associated with a higher inclination to get a coronavirus vaccine, exposure to vaccine misinformation lowered that inclination by several percentage points (Loomba et al. 2020).

All of this suggests that communications has an important role to play in the vaccination phase. However, the research summarised here shows that no one media source or actor can expect to reach everyone, so communication strategies should consider a range of platforms and sources. The infodemically vulnerable that do not use the news media for their coronavirus information might turn to social media instead, so attempts to combat misinformation on these platforms will be critical, not just for this group but for everyone because of their huge reach. And while news media seem to serve older, more affluent consumers well, they currently struggle to reach many less privileged parts of the population.

Finally, as we wrote in our final report for the UK COVID-19 News and Information Project: “In terms of the broad public, the simplest suggestion for communications in the next stages of the coronavirus crisis is to focus less on politicians and pundits, except where absolutely necessary, and more on the sources that are (a) highly and broadly trusted and (b) demonstrably help people understand the crisis, most notably the NHS and scientists, doctors, and other experts” (Nielsen, Fletcher, Kalogeropoulos, et al. 2020).