Communications in the coronavirus crisis: lessons for the second wave

People walk past a signage about COVID-19 measures as the spread of the coronavirus disease (COVID-19) continues in Slough, Britain October 23, 2020. REUTERS/Paul Childs

People walk past a signage about COVID-19 measures in Slough, Britain. REUTERS/Paul Childs

This report is part of the UK COVID-19 news and information project.

DOI: 10.60625/risj-0666-xv66

Introduction 

The UK has suffered through one of the worst coronavirus epidemics in the world, with more than 650,000 confirmed cases by mid-October, over 43,000 casualties across the country, and disadvantaged, marginalised, and vulnerable communities particularly hard hit.

Over the summer, the daily number of new cases and casualties was greatly reduced after a stringent lockdown in the spring, but the crisis’s often severe knock-on consequences for education, the economy, mental health, and other areas have caused growing concern, and the question of how to balance different priorities has become explicitly political and often controversial.

This represents a communications crisis as well as a public health crisis, and understanding the role of communications, news, and media in the handling of the epidemic itself as well as its wider social and political impact requires attention to how people navigate the crisis, something we have been investigating since March (Nielsen et al. 2020a). Our work has documented how the situation has changed dramatically in just a few months. After an initial surge in news use, news consumption in the UK has gradually returned to pre-crisis levels, news avoidance has grown, and trust in key sources of COVID-19 news and information has declined (Nielsen et al. 2020b). Digital platforms, including social media, video sharing sites, messaging applications, and search engines have seen high levels of use throughout the crisis, and often promote official health communication, but have also had serious problems with misinformation, and few trust them for information about the coronavirus (Nielsen et al. 2020c). We have also seen a dramatic decline in public trust in the UK government as a source of information about COVID-19 and a significant increase in the number of people who see the UK government itself as a source of potentially false or misleading information about the coronavirus (Fletcher et al. 2020a).


Listen to this episode about the report

Listen to the episode: On Apple Podcast | On Spotifyon Google Podcasts


While the situation on the ground in some ways improved over the summer, trust in news, trust in information from platforms, and trust in the government did not rebound. Even before the second wave started in late August, a majority of the public continued to see the coronavirus crisis as the most important issue facing the UK, though the political focus shifted from public health to boosting the economy and reopening after the lockdown (Nielsen et al. 2020b).

With the second wave now well underway and Prime Minister Boris Johnson warning of a very tough winter ahead, the situation in the UK remains challenging and precarious, in terms of the crisis broadly, and in terms of coronavirus communications more specifically. The daily number of new confirmed cases is currently increasing, there are continued problems with both testing and tracing, and the UK government’s handling of the crisis is under intense scrutiny from news media, scientists, the Labour opposition, the Conservative back benches, political leaders representing the devolved nations and hard-hit areas in England, as well as by the wider public. The early ‘rally around the flag’ effect that led 72% to say that the UK government was handling the issue of coronavirus ‘very’ or ‘somewhat’ well in late March has dissipated and, after months of decline, the figure hit 31% by late September.1

Handling the second wave and the months ahead is about getting the public health response right, but it is also about handling the coronavirus communications crisis. In the UK, our research shows that this part of the crisis is characterised by a combination of a public that mostly comes across as informed, cautious, and willing (at least in theory) to take additional precautions, but also growing information inequality, eroding trust, and a growing and significant minority of ‘infodemically vulnerable’ who make little or no use of news about the pandemic and do not trust the news media. ‘Government by communication’ is a central part of handling the crisis, as new rules, regulations, and other formal measures take time to put in place. But government by communication grows harder as fewer people follow the news or trust the government, and when many feel neither the news media nor the government are helping them navigate the crisis and how to respond to it.

In this Reuters Institute report, we look at three key lessons learned on communications in the coronavirus crisis and look to the months ahead. We focus on communications because communication is central to any crisis, including a public health crisis, and is central to the political discussion around how we, as a society, handle them. Information from a wide range of sources, as well as people’s perception of the trustworthiness of these sources, will influence how they understand and respond to the crisis, and how they evaluate which institutions are helping address it (and which ones not). As researchers have long known, it is perceptions of risk, not actual risk, that determine how people respond to crises (Glik 2007), and these perceptions are influenced in large part by information from news organisations, sometimes by misinformation and disinformation, and by many other sources going well beyond official communication by governments and public health authorities. This is why, as WHO Director-General Tedros Adhanom Ghebreyesus said in February, with the arrival and spread of COVID-19, ‘we’re not just fighting an epidemic; we’re fighting an infodemic’, a deluge of information, some of which is misinformation, political propaganda, rumours, or other forms of unreliable material.2

From a public health perspective, the UK may face the second wave in some ways better equipped to deal with the epidemic. But the erosion in trust in key institutions we saw in the spring and summer means it is less well equipped to deal with the coronavirus communications crisis. Doing so effectively with waning attention and trust will require learning from the spring and summer and special emphasis on engaging those most at risk. To help with that, we offer three lessons identified on the basis of our work on the UK COVID-19 news and information project, where we have worked to analyse the role of news and media in the crisis over the last six months.

Key Findings

Based on analysis of data from ten waves of an online panel survey with the same respondents surveyed at regular two-week intervals from mid-April till mid-August 2020, we find that:

  • With some exceptions, most of the UK public are informed about COVID-19 as a disease, report that they have behaved cautiously and mostly followed government guidelines, and say they are willing to take precautionary measures if instructed to do so.
  • However, news use has declined during the crisis after the initial surge, trust in news has fallen, trust in the government as a source of information about COVID-19 has dropped dramatically, and as the UK faces the second wave, a large minority of the public – an estimated 20 million people – do not feel that the news media and/or the government have explained what they can do in response to the pandemic.
  • Furthermore, information inequality is a real and growing problem, with systematic inequalities around age, gender, as well as income and education in how people engage with information about the coronavirus, suggesting that the ways people navigate the second wave and make sense of the far more explicitly politicised and often polarising responses to it will be even more marked by inequality than earlier parts of the crisis.
  • Offering a preliminary definition of the ‘infodemically vulnerable’ as the subset of the public who consume little to no news and information about COVID-19, and say they would not trust it even if they did, we can provide a first estimate of the size of this group, which has grown from a small minority of 6% early in the crisis to a significantly larger minority of 15% by late August – an estimated 8 million people who are more at risk of being at best less informed and at worst un- or misinformed.
  • 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.

Methodology 

The data we use for this report come from the UK COVID-19 news and information project. The project analyses how the British public navigates information and misinformation about coronavirus and how the government and other institutions are responding to the pandemic. It is based on a ten-wave online panel survey of a representative sample of the UK population, and data on news supply from the most popular UK news outlets.

The survey was designed by the Reuters Institute for the Study of Journalism to collect data on how people navigate news and information during the coronavirus pandemic. The survey was fielded online by YouGov. Starting in April 2020, ten waves of the survey were fielded at two-week intervals. The survey is a mix of tracking questions and specific questions fielded only in some waves.

Waves and sample sizes

  • 1. 10–14 April 2020 = 2,823
  • 2. 24–28 April 2020 = 2,291
  • 3. 7–13 May 2020 = 1,973
  • 4. 21–27 May 2020 = 1,774
  • 5. 4–10 June 2020 = 1,645
  • 6. 18–24 June 2020 = 1,467
  • 7. 2–8 July 2020 = 1,338
  • 8. 16–22 July 2020 = 1,218
  • 9. 30 July–5 August 2020 = 1,117
  • 10. 13–19 August 2020 = 1,003

As this is a panel survey, we attempt to survey the same people in every wave. However, for a variety of reasons, some people do not complete every survey. Those that do not complete the most recent survey are not invited to respond in the next wave, meaning that the sample for each wave only consists of people that completed every previous survey. Because panel attrition is non-random (our panel contained more older people and more women by Wave 10), we separately weight each wave by age, gender, region, education, and social grade so that we have a nationally representative sample in each.

We should note that online samples will tend to under-represent the consumption habits of people who are not online (typically older, less affluent, and with limited formal education). It is also important to note that surveys rely on recall, which is often imperfect or subject to biases. We have tried to mitigate these risks through careful questionnaire design and testing. On the other hand, surveys can be a good way of capturing fragmented media consumption across platforms (e.g. social media, messaging, apps, and websites), and tracking activities and changes over time, and provide representative samples that allow for generalisation.

Where we refer to differences between groups, these are statistically significant using a z-test for population proportions (p < .05).

More information about the project 

First lesson 

Most people are relatively informed but large minority do not feel news media or the government have explained what they can do in response to the pandemic

With some exceptions, our findings suggest that most of the UK public are informed about COVID-19 as a disease, report that they have behaved cautiously and mostly followed government guidelines, and say they are willing to take precautionary measures if instructed to do so.

First, most of the public answer correctly when we ask them basic factual questions about COVID-19 as a disease, in particular the steps they can take to protect themselves and some of the policy issues around it. To gauge how informed people are, in August we fielded eight different factual questions with one factually correct option, several alternatives, as well as a ‘don’t know’ option. The point of a scale like this is not to ask an exhaustive list of all possible questions, or even analyse responses to individual ones in isolation, but to be able to get an overall sense of how informed people are across a range of different relevant issues. The central finding here is that the large majority of respondents (75%) in late August answer the majority of the questions (five or more) correctly.

Furthermore, many respondents give answers that are incorrect but arguably unlikely to leave them worse off, because they err on the side of caution – e.g. 85% give the correct answer to the question ‘according to the NHS, how long should you wash your hands for in order to protect yourself from coronavirus’ (20 seconds), but among those who do not, just 2% say ‘don’t know’ and 1% ‘10 seconds’, while the remaining 12% answer 30 or 40 seconds. Similarly, at a first glance, it is concerning that just 26% give the correct answer to the question ‘According to the UK government, how long should you stay at home for if you have symptoms of coronavirus?’ (10 days). It is perhaps less concerning in light of the fact that 70% answer 14 or 21 days. Similar patterns are seen for questions including what percentage alcohol content hand sanitiser is effective against the virus, the average incubation time, etc. If we are generally better off safe than sorry, these answers are reassuring. We also find that specialised jargon presumably unknown to most people in February is by now familiar to most – including ‘antibody test’ (77% answer the relevant question correctly) and ‘R0’ (86%). Judged on their responses to our questions, most people come across as well informed about the coronavirus.

According to the NHS, how long should you wash your hands for in order to protect yourself from coronavirus?

  • 10 seconds = 1%
  • 20 seconds (correct) = 85%
  • 30 seconds = 7%
  • 40 seconds = 5%
  • Don’t know = 2%


What do you think the R0 (R nought) number refers to?

  • The average number of people that an infected person will spread a virus to (correct) = 86%
  • The average number of people that will die if they catch a virus = 2%
  • The average number of people tested for a virus = 2%
  • The average number of people that become seriously ill they catch a virus = 2%
  • Don’t know = 8%


According to the UK government, how long should you stay at home for if you have symptoms of coronavirus?

  • 3 days = 0%
  • 10 days (correct) = 26%
  • 14 days = 68%
  • 21 days = 2%
  • Don’t know = 3%


According to the World Health Organisation (WHO), what is the average time between when people catch coronavirus and when they start showing symptoms?

  • 5-6 days (correct) = 43%
  • 10-12 days = 45%
  • 20-25 days = 3%
  • 26-30 days = 1%
  • Don’t know = 8%


Which country halted its funding to the World Health Organisation (WHO) in April 2020?

  • China = 2%
  • USA (correct) = 87%
  • UK = 2%
  • Russia = 1%
  • Don’t know = 8%


Which of the following has adopted a much less strict coronavirus ‘lockdown’ than most other countries in Europe?

  • Germany = 2%
  • Spain = 1%
  • Sweden (correct) = 82%
  • UK = 6%
  • Don’t know = 9%


What is a coronavirus antibody test?

  • A test for whether you currently have coronavirus = 7%
  • A test for whether you are likely to become seriously ill from catching coronavirus = 1%
  • A test for whether you are immune to coronavirus = 11%
  • A test for whether you have had coronavirus in the past (correct) = 77%
  • Don’t know = 4%


Hand sanitiser needs to contain what percentage of alcohol content in order to be effective against coronavirus?

  • At least 20% = 4%
  • At least 40% = 5%
  • At least 60% (correct) = 48%
  • At least 80% = 34%
  • Don’t know = 10%

NB: numbers don’t always add up to 100% because of rounding

Second, in addition to being informed, most of the public also appears cautious. We should stress that surveys which rely on self-reported answers are not ideal sources of data on behaviour, in particular in areas where social norms and personal interests may be at odds, or when people are confronted with the actual implications of strict compliance, for example when told to self-isolate (Smith et al. 2020). So our data on what people say they do is thus at best only indicative. But it is still worth reporting that when asked to describe their behaviour during the pandemic, people report behaving cautiously. Almost half of our respondents insist that they are always following government guidelines, and taking measures like staying at home, working from home, limiting contact with others, keeping 2 metres distance, and washing their hands regularly (Figure 1).

If we include those who answer ‘most of the time’, 90% or more say they follow most of these guidelines and recommendations, and the main variations are around social inequality, e.g. those with low income and low levels of formal education are much less likely to say they are working from home than those with high income and high education, reflecting different kinds of jobs. As said, this is not hard proof that people have always in fact been quite so compliant in every aspect and every situation. But the self-reported data are encouraging, and the variation in the anonymous responses well enough aligned with expected differences to suggest that they capture at least people’s desire to take action where their circumstances allow it. (If respondents were simply driven by social desirability bias it is hard to see why so many relatively less privileged people would report they still go to work – something also found by e.g. Safi et al. 2020.)

Figure 1.

Third, the public also seems receptive to suggestions as to how they can protect themselves, their loved ones, and their communities. On most of the additional precautionary measures for which we have data, large majorities say they would be willing to take these if advised to (Figure 2). (Again, these are self-reported data, and in this case hypothetical questions, but still indicative of public opinion.) Between 75% and 90% of respondents say they definitely or probably would take most of the preventive measures we asked about. (The main outlier is whether people would be willing to download the contact tracing app, which in August 28% said they definitely would, and another 22% they probably would.) For the minority who are not in advance ready to say they will take additional preventive measures, it is a mixed group. On most individual questions, the group who answer ‘maybe’ equals or exceeds the small minority who answer ‘probably’ or ‘definitely’ not.

Figure 2.

So overall, our data suggest the UK public is informed, cautious, and open to suggestions on how they can protect themselves, their loved ones, and their communities from COVID-19.

Are these suggestions coming across? Not to everyone. As of August, 61% say the news media have ‘explained what I can do in response to the pandemic’ (down from 73% in April), and 58% say the same about the government (down from 67% in April). This means about 20 million people out of the UK adult population of about 52 million do not feel that the news media have explained what they can do in response to the pandemic.3 The same is true for the government. Despite untold news articles published and countless government press conferences, statements, speeches, and ads across online platforms and offline media, millions of people across the UK are fundamentally unsure about what to do in this difficult situation.

This is an example of how the coronavirus crisis is also a communications crisis. The news media and the government may feel they have explained what people can and should do. But millions of people, even people who are otherwise informed, cautious, and open to suggestions, do not see it that way.

Second lesson 

Information inequality is growing as the crisis continues

While the early stages of the coronavirus crisis saw both a ‘rally around the flag’ effect, contributing to a brief period of high trust in the government, and a parallel ‘rally around the news’ effect, as people came together around widely used and broadly trusted news media, much of this had evaporated by late June (Fletcher et al. 2020b).

The big initial surge in news use was followed by a slow but consistent decline, and as news use fell, inequalities in COVID-19 news use grew, and news use became more unevenly distributed. This is an important facet of the coronavirus communication crisis, because relying on information about the coronavirus from news organisations – with all their imperfections – is significantly and positively associated with being better informed (Nielsen et al. 2020a), and because news media normally provide more accessible, timely, and independent information than most other sources.

Growing inequality in COVID-19 news use in part reflects people’s own choices and interests, but it is important to recognise that these choices and interests are often partially aligned with various forms of structural inequality (as also found by e.g. Nguyen et al. 2020). During the first three months, gaps in news use by age grew, as did differences by gender, and differences in levels of COVID-19 news use by household income and levels of formal education remained significant and stable over time. News avoidance also rose slightly, and has remained consistent since late May, with more than one in five saying they always or often actively avoid the news, and women consistently more likely to say they avoid news than men (Fletcher et al. 2020c).

The systematic inequalities around age, gender, and both income and education in how people engage with information about the coronavirus have remained broadly stable in the months since, suggesting that the ways people navigate the second wave and the months ahead will be even more marked by inequality than earlier parts of the crisis.4

First, differences by age (Figure 3). At the start of our panel survey in mid-April we saw a difference in COVID-19 news use between age groups. Older people (categorised as those aged 55 and over at the start of the study) used more news than younger adults (those aged 18–54) with a 12 percentage point (pp) gap between the 86% of those aged 55 or over saying they accessed COVID-19 news at least once a day on average, compared to 74% of those aged 18–54.5 By the end of June, the gap had doubled to 24pp (75% vs 51%), as news use fell overall, but more sharply within the younger age group. By the end of August, the gap was still 21pp (67% vs 46%).

Figure 3.

Second, differences by gender (Figure 4). In April, roughly equal proportions of both men (78%) and women (79%) said they were accessing COVID-19 news at least once a day on average. By late June, an 8pp gap had emerged, with women less likely to regularly access COVID-19 news than men. As noted before, this is in line with previous research documenting that men often consume more news than women, and an illustration of how the dissipation of the ‘rally around the news’ and surge in news use early in the crisis comes with a return to longstanding structural inequalities in news use, perhaps exacerbated in cases where the pandemic has reinforced existing inequalities, e.g. in who takes responsibility for caregiving (Toff and Palmer 2019). By the end of August, the gap was still 6pp.

Figure 4.

Third, differences associated with indicators of social class, such as levels of formal education (Figure 5) or household income (Figure 6). Here we find that, despite the early surge in news use, those with lower levels of education and household income have been consistently less likely to consume news about COVID-19 once a day or more on average throughout much of the crisis.6 Overall, news use has declined in all groups, and the gaps between different levels of income and education have remained roughly the same. This is a reminder that even in a media environment where news is at least in principle easily accessible, where public service media have an obligation to serve all audiences and popular newspapers a long tradition of successfully reaching a wide audience, and where the situation suggests news media could really help people navigate a difficult situation, less privileged parts of the population are significantly less likely to turn to news media than their more privileged counterparts are.

Figure 5.

Figure 6.

Marked and often growing differences in how actively people seek out news and information about COVID-19 does not change the overall picture documented in the first part of this report – that much of the public is informed, cautious, and open to suggestions on what to do. As we have shown elsewhere (see Fletcher et al. 2020b), even those who are more casual news users and those who distrust the government continue to be relatively well informed about the crisis.

But if sustained over time, information inequality will mean that different parts of the public have different levels of understanding both of the public health situation overall, the most up-to-date understanding of the coronavirus as a disease, what they are meant to do to protect themselves, and of how the government and other institutions are responding to the crisis. Official rules and guidelines are, after all, complex, vary across the country, and keep changing – in a situation where some cabinet ministers and even the Prime Minister have on occasion struggled to remember, let alone explain, what the rules are, those whom news media fail to reach will be less able to stay on top of the situation. Growing information inequality may also potentially increase the UK’s vulnerability to various kinds of misinformation if news media are unable to reach growing parts of the public, as existing research indicates that news media, with their various imperfections, can help increase societies’ resilience to online disinformation (see e.g. Humprecht et al. 2020, in line with earlier research on how media can inform democracies, see e.g. Aalberg and Curran 2012).

The growing information inequality we document is associated with structural inequalities along lines of age, gender, class, and similar differences almost certainly exist around other forms of structural inequality too. These inequalities shape people’s sense of whether the news media have explained what they can do in response to the pandemic – older people, those with high income, and degree holders are more likely to say they have done this. These inequalities will shape how the public navigates the second wave and makes sense of the ongoing coronavirus crisis in the UK.

Third lesson 

The ‘infodemically vulnerable’ are a small but significant and growing part of the UK public

While the majority of the UK public, with some variation, are informed and continue to follow the news to navigate the crisis, even as information inequality is growing along lines of age and gender, a small but significant and growing minority is at risk of being left (at best) less informed and (at worst) misinformed.

As already mentioned, one of the key trends to emerge during the first wave of coronavirus in the UK was the decline in trust for news organisations as a source of information about COVID-19 – falling from 57% of the population in April to 45% in mid-August, during a period in which trust in the government declined dramatically and trust in COVID-19 information from various digital platforms remained low (Figure 7).

Figure 7.

COVID-19 news use also fell during the same period. In our first report, we documented a rally around the news effect where use surged in April as the UK went into lockdown, but started to decline soon after (Fletcher et al. 2020b). The proportion who got news about COVID-19 at least once a day per week on average dropped by 24pp from 79% in mid-April to 55% in mid-August (Figure 8).

Figure 8.

Each of these two declines, in trust in news, and in coronavirus news use, could be seen as a cause for concern on their own, but in combination they suggest a more specific problem: the emergence of an ‘infodemically vulnerable’ group who both consume little to no news and information about COVID-19, and do not trust news and information about COVID-19 even if it reaches them. We define this group by analogy to medical research identifying epidemiologically vulnerable groups on the basis of underlying health conditions and public health research identifying social vulnerability on the basis of structural inequality. Just as the epidemiologically vulnerable are not necessarily ill or going to fall ill, but are more at risk, the infodemically vulnerable are not necessarily mis- or uninformed, but more at risk.

For the purpose of this report, we define the infodemically vulnerable as those who (a) consume little to no news about COVID-19 from news organisations and (b) have low trust in COVID-19 information from news organisations. This preliminary definition draws on our own work documenting how following the news helps people stay more informed about the coronavirus (Nielsen et al. 2020a, in line with a broader body of literature on how news use helps people stay informed, see e.g. Aalberg and Curran 2012) as well as existing research (e.g. Larson 2020) on health communication, arguing that it is often when authoritative information sources are perceived as untrustworthy that the climate is set for the viral spread of unfounded speculation online and offline (whether driven by ordinary people acting in good faith, by committed activists, some media, or some politicians with more ambiguous motives, or a combination thereof). The combination of these two factors leaves people at risk of being infodemically vulnerable. Those who follow the news, even if they do not trust it, can still benefit from the information provided. Those who do not follow the news but trust it can, in most cases, easily access relevant information if and when they feel they need it. But those who neither follow the news nor trust it are arguably more at risk – they are less likely to receive accessible, timely independent information about the coronavirus, and potentially less likely to act upon it even if it does reach them, more at risk of being less informed (not necessarily misinformed), and less equipped to counter misinformation should they be exposed to it, for example, by relying on unreliable sources for COVID-19 advice and information.7 Previous research has suggested that identifiable misinformation is a very small part of most people’s overall information intake – little more than 1% of Americans’ daily news consumption according to one study (Allen et al. 2020) – but may be a greater risk to those who consume very little news and don’t trust news but still come across at least some misinformation.

Our definition of the infodemically vulnerable allows us to estimate the size of the group in the UK and track how it has changed over time, by identifying the subset of our respondents in each wave of our survey who (i) say they consumed COVID-19 news less than once a day on average from news organisations (via TV, radio, print, or online), and (ii) say they have low trust in COVID-19 news from news organisations.8 We recognise that this is a limited and preliminary definition, but it allows us to identify a small but significant part of the public whose information needs are not being met and who are therefore more at risk.

Figure 9.

Our analysis suggests that the infodemically vulnerable group in the UK grew from 6% of the population in mid-April, rising to roughly 14% in late June, and stayed at this level through to August where it hit 15% in the last wave of our survey (Figure 9). This is a far bigger group than those who believe in individual conspiracy theories like the non-existent link between 5G and COVID-19 (2% in our last wave), and reveals a more systemic problem.

The group is roughly evenly split by gender, but there are differences by age and education. By the final wave of the survey (13–19 August), those under 35 were more likely to be in the infodemically vulnerable group (20%) than those aged 35 and over (14%). Similarly, those that were educated to degree level (11%) were less likely to be infodemically vulnerable than those that were not (18%). The infodemically vulnerable group is thus characterised by some, but not all, of the same structural inequalities we have documented around news use more broadly.

The infodemically vulnerable represent a small but significant and growing part of the population more at risk of being (at best) less informed than the public at large, and (at worst) being uninformed and more susceptible to outright misinformation. As the UK navigates the second wave in the months ahead, it is important to recognise that this group of people who are considerably less likely to receive news and information about COVID-19, and less likely to act upon it should it reach them, has grown to 15%. This is roughly 8 million people out of the adult UK population of about 52 million.9 During the early ‘rally around the news’, the group was much smaller, at 6%. It is possible that later stages in the crisis will see a resurgence in news use, and perhaps even trust in news. But in the course of the six months for which we have data, news use and trust in news have been in constant decline overall. Unless this changes, the UK will be less well positioned to deal with the coronavirus communication crisis in the months ahead, in part because it has a much larger minority of infodemically vulnerable people than earlier in the crisis.

Conclusion

The coronavirus communications crisis in the UK is characterised by declining news use, eroding trust in news, low trust in platforms, and a dramatic decline in trust in the UK government as a source of information about COVID-19, combined with growing information inequalities associated with age and gender, and a small but significant and growing part of the public who are ‘infodemically vulnerable’. Most members of the public are informed, say they are behaving cautiously, and that they are willing to take additional precautions, but millions of people do not feel the news media or the government have explained what they can do, do not trust them, and do not pay them much attention.

Are there, as the Prime Minister has argued in Parliament, in the wider public a few who are ‘brazenly defying the rules’ and otherwise putting themselves and others at risk? Undoubtedly. But overall our findings provide little support for attempts to suggest that the UK public is somehow responsible for the fact that the UK is suffering one of the worst coronavirus crises in the world. While our results do not support claims by some politicians and some news media that seem to blame the public, we can document that much of the UK public is in effect returning the favour and blaming politicians and news media for their handling of the crisis. It is not simply that 43% of respondents in the last wave of our survey in August said that the government was doing a bad job responding to the coronavirus pandemic, and 38% were very or extremely concerned about false or misleading information from the UK government about coronavirus (these numbers are from our survey research, similar concerns have been identified in qualitative research, see e.g. Kyriakidou et al. 2020). We also found that 35% felt that the coronavirus situation has been made worse by how the news media have covered it, just as trust is very low in information about COVID-19 found via search engines and, especially, via social media, video sharing sites, and messaging applications (for more on this see Nielsen et al 2020c). 

Thus while the UK in some ways may face the second wave and the winter ahead at least somewhat better equipped to deal with the epidemic from a public health perspective, the erosion of trust in government and news media, increasing information inequality, and a growing minority of infodemically vulnerable people means that it is in many ways less well equipped to deal with the coronavirus communications crisis. Communication was hard enough in the spring when there was more attention, more trust, and greater political consensus on most major decisions. It will be harder, much harder, in a situation characterised by waning attention, low trust, and increasingly explicit political disagreement and polarisation. Much of the British public feels key institutions have failed in the first part of the crisis. This will shape how people navigate the rest of the crisis. And this will in turn shape how well the UK, as a society, is able to handle that crisis.

How might the UK find a way through the coronavirus communications crisis in light of these three lessons from the spring and summer? If the second wave continues to get worse, not just in terms of the daily number of confirmed COVID-19 cases, but also in terms of deaths, and in terms of the impact on daily life, local communities, and the economy, there may be another surge in news use. It is harder to see how there could be a repeat of the ‘rally around the flag’ and ‘rally around the news’ seen earlier in the crisis, as trust has consistently declined and many of the issues at hand have become increasingly explicitly politicised and attitudes to them more polarised. Perhaps trust, as the saying goes, arrives on foot and leaves on horseback.

This complicates both ‘government by communication’ and news coverage, necessary parts of helping the whole UK through the crisis. Some of the complications ahead are arguably an inevitable consequence of the crisis becoming more explicitly political in ways that makes public health communication harder but are nonetheless part and parcel of how any democracy deals with difficult, complex, and enormously important decisions. This is politics, and difficult as it may be, it cannot – and in our view should not – be made to go away, no matter how much one may or may not be following the science.

But the findings from our research also provide some indications of what can be done, beyond learning to live with more explicit disagreement and beyond wider efforts to respond to misinformation online and limit its pernicious influence on public attitudes and behaviours (see e.g. Donovan 2020, Palen et al. 2020, Vraga and Bode 2020, for suggestions on this).

The infodemically vulnerable group is more likely to use social media such as Facebook and Twitter, and even though they, by definition, use all news sources less, the BBC is still their most widely used source of news. Most people say they do not trust news and information about COVID-19 they come across on social media, but even with that in mind, the measures platforms have taken, such as the dedicated COVID-19 tab in Twitter’s Explore and Facebook’s COVID-19 Info Center, as well as the free ads some of these companies have offered health authorities, may be useful in reaching the infodemically vulnerable. Challenging as it is to reach infrequent news users who do not trust news media, the BBC is responsible for serving all audiences, and doing this is arguably especially urgent in a crisis like this.

For the broader public who are informed, cautious, and open to suggestions, but in many cases no longer trust the government or the news media, let alone digital platforms, as sources of information about the coronavirus, the simplest suggestion might be: more doctors, nurses, and scientists, and fewer politicians and pundits. Foregrounding highly and broadly trusted expert sources who demonstrably help people understand the crisis could help enhance public engagement, understanding, and trust overall. It might even help reach the infodemically vulnerable, who trust the NHS and various experts less than do the public at large, but far more than they trust news organisations or the government.

Part of the situation we analyse here is a crisis of trust more than it is about an absence of information (though there are things we do not know, or where our knowledge is evolving) or even necessarily about misinformation (as real a problem as this is). But it is important to recognise that, while trust in many institutions has declined, some stand out and remain highly and broadly trusted. Trust in the UK government has declined by 22pp since our first survey wave in April to 45%, and trust in news organisations by 12pp, also to 45%. But the NHS, scientists, doctors, and other experts, and global health organisations like the WHO, are very highly and broadly trusted, and have seen smaller declines, with trust figures by late August of, respectively, 85%, 82%, and 75%. Not only are the NHS and scientists highly and broadly trusted, we also find in our earlier research that those who say they rely on them for information about coronavirus are more informed (Nielsen et al. 2020a).

So it is not trust overall that has collapsed during the crisis. It is, first, trust in news that has fallen and, second, trust in the government that has declined dramatically. In light of this, those interested in helping the public navigate the crisis in the months ahead would do well to foreground those the public still trusts, and who also have proven expertise and demonstrably help people understand the crisis. News coverage has sometimes suggested that ministers need to ‘balance science and politics’, but attention has often focused more on the politics and less on the science, just as the gap between the government’s line and the recommendation of its own scientific advisers has grown. Scientists and health authorities have no right to make these big important decisions on our behalf – in a democracy only the public and its elected representatives do. But experts can inform our decisions. With that in mind, perhaps it is time to reconsider what balance serves the public best, perhaps in policy, but also in how we communicate about the coronavirus crisis.

Footnotes

1 https://yougov.co.uk/topics/international/articles-reports/2020/03/17/perception-government-handling-covid-19

2 https://www.who.int/dg/speeches/detail/munich-security-conference

3 “The 20 million figure is a rough estimate based on the latest ONS data on the size of the 18+ population (https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/articles/overviewoftheukpopulation/august2019) and data from our survey. The 20 million figure is indicative and should be treated with caution due to the uncertainties associated with extrapolating from responses to an online survey to population level.”

4 We look here only at structural inequalities associated with age, gender, and class indicators like education and income, but it is important to recognise that there are many troubling examples of other forms of structural inequality, including e.g. race and ethnicity, shaping how the crisis impacts different groups, often compounding intersecting inequalities. Unfortunately, our number of respondents from individual black and minority ethnic groups – particularly in later survey waves – are too small to allow for robust statistical analysis, and we therefore have to leave this important dimension of the crisis aside.

5 In the survey we asked people ‘On how many of the last 7 days have you used each of the following as a source of news and information about coronavirus (COVID-19)?’ Respondents could select a number between 0 to 7 days for ‘Television news bulletins or programmes such as News at Ten, C4 News, Good Morning Britain, Newsnight and Question Time’, ‘24 hour news television channels such as Sky News or BBC News 24’, ‘Radio news bulletins or programmes such as BBC Today Programme, BBC 5 Live, LBC, commercial radio bulletins’, ‘Printed newspapers such as The Guardian, Times, Daily Mail, Mirror’, ‘Printed magazines such as the Economist or The Week’, ‘Websites/apps of newspapers such as Guardian online, Times online, Mail Online’, ‘Websites/apps of news magazines such as The Economist or The Week Online’, ‘Websites/apps of TV and Radio companies such as BBC News Online or Sky News Online’, and ‘Websites/apps of other news outlets such as MSN, Yahoo, Huffington Post, BuzzFeed, Vice News’. These numbers were summed for each respondent, with those less than 7 placed in the ‘less than once a day on average’ category, and those with 7 or more placed in the ‘once a day or more on average’ category.

6 We group respondents into ‘High’, ‘Medium’, and ‘Low’ household income categories. We classify low household income as £0–19,999 per year, medium as £20,000–44,999, and high as £45,000 and higher. Around 18% of respondents ‘Prefer not to say’ when asked about their household income, and these respondents are excluded from the analysis.

7 News use and trust in news are of course not the only factors that can leave people infodemically vulnerable and thus particularly susceptible to the negative consequences of misinformation about coronavirus. Social factors will matter too (e.g. recent research from the US has shown that young people are more likely to believe false information about coronavirus (Baum et al. 2020)) as do political factors (some studies highlight that those with strong partisan preferences are more susceptible to health misinformation: Scherer and Pennycook 2020).

8 The measure of COVID-19 news use frequency is described in footnote 4. Respondents with low trust in COVID-19 news from news organisations are those that selected between 0 and 5 on the 0–10 trust scale.

9 The 8 million figure is a rough estimate based on the latest ONS data on the size of the 18+ population (https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/articles/overviewoftheukpopulation/august2019) and data from our survey. The 8 million figure is indicative and should be treated with caution due to the uncertainties associated with extrapolating from responses to an online survey to population level.

References

  • Aalberg, T., Curran, J. (eds). 2012. How Media Inform Democracy: A Comparative Approach. New York: Routledge.
  • Allen, J., Howland, B., Mobius, M., Rothschild, D., Watts, D. J. 2020. ‘Evaluating the Fake News Problem at the Scale of the Information Ecosystem’, Science Advances 6(14), 1–6.
  • Baum, M. A., Ognyanova, K., Chwe, H., Quintana, A., Perlis, R. H., et al. 2020. The State of the Nation: A 50-State COVID-19 Survey: Report #14: Misinformation and Vaccine Acceptance.
  • Donovan, J. 2020. ‘Concrete Recommendations for Cutting through Misinformation during the COVID-19 Pandemic’, American Journal of Public Health 110(S3), S286–7.
  • Fletcher, R., Kalogeropoulos, A., Nielsen, R. K. 2020a. Trust in UK Government and News Media COVID-19 Information Down, Concerns over Misinformation from Government and Politicians Up. Oxford: Reuters Institute for the Study of Journalism.
  • Fletcher, R., Kalogeropoulos, A., Simon, F. M., Nielsen, R. K. 2020b. Information Inequality in the UK Coronavirus Communications Crisis. Oxford: Reuters Institute for the Study of Journalism.
  • Fletcher, R., Kalogeropoulos, A., Nielsen, R. K. 2020c. News Avoidance in the UK Remains High as Lockdown Restrictions are Eased. Oxford: Reuters Institute for the Study of Journalism.
  • Glik, D. C. 2007. ‘Risk Communication for Public Health Emergencies’, Annual Review of Public Health 28(1), 33–54.
  • Humprecht, E., Esser, F., Van Aelst, P. 2020. ‘Resilience to Online Disinformation: A Framework for Cross-National Comparative Research’, International Journal of Press/Politics 25(3), 493–516.
  • Kyriakidou, M., Morani, M., Soo, N., Cushion, S. 2020. ‘Government and Media Misinformation about COVID-19 is Confusing the Public’, LSE Blogs.
  • Larson, H. J. 2020. Stuck: How Vaccine Rumors Start – and Why they don’t Go Away. New York: Oxford University Press.
  • Nguyen, M. H., Gruber, J., Fuchs, J., Marler, W., Hunsaker, A., Hargittai, E. 2020. ‘Changes in Digital Communication during the COVID-19 Global Pandemic: Implications for Digital Inequality and Future Research’, Social Media + Society, July–September, 1–6.
  • Nielsen, R. K., Fletcher, R., Newman, N., Brennen, J. S., Howard, P. N. 2020a. Navigating the ‘Infodemic’: How People in Six Countries Access and Rate News and Information about Coronavirus. Oxford: Reuters Institute for the Study of Journalism.
  • Nielsen, R. K., Kalogeropoulos, A., Fletcher, R. 2020b. Most in the UK Say News Media have Helped them Respond to COVID-19, But a Third Say News Coverage has Made the Crisis Worse. Oxford: Reuters Institute for the Study of Journalism.
  • Nielsen, R. K., Kalogeropoulos, A., Fletcher, R. 2020c. Social Media Very Widely Used, Use for News and Information about COVID-19 Declining, Oxford: Reuters Institute for the Study of Journalism.
  • Palen, L., Anderson, J., Bica, M., Castillos, C., Crowley, J. et al. 2020. Crisis Informatics: Human-Centered Research on Tech & Crises: A Guided Bibliography Developed by Crisis Informatics Researchers.
  • Safi, M., Coulangeon, P., Godechot, O., Helmeid, E., Pauly, S. et al. 2020. When Life Revolves around the Home: Work and Sociability during the Lockdown.
  • Scherer, L. D., Pennycook, G. 2020. ‘Who is Susceptible to Online Health Misinformation?’, American Journal of Public Health 110(3), S276–7.
  • Smith, L. E., Potts, H. W. W., Amlot, R., Fear, N. T., Michie, S., Rubin, J. 2020. ‘Adherence to the Test, Trace and Isolate System: Results from a Time Series of 21 Nationally Representative Surveys in the UK (the COVID-19 Rapid Survey of Adherence to Interventions and Responses [CORSAIR] Study).’
  • Toff, B., Palmer, R. A. 2019. ‘Explaining the Gender Gap in News Avoidance: “News-Is-for-Men” Perceptions and the Burdens of Caretaking’, Journalism Studies 20(11), 1563–79.
  • Vraga, E. K., Bode, L. 2020. ‘Correction as a Solution for Health Misinformation on Social Media’, American Journal of Public Health 110(S3), S278–80.

 

About the Authors

Rasmus Kleis Nielsen is the Director of the Reuters Institute for the Study of Journalism and Professor of Political Communication at the University of Oxford.

Richard Fletcher is a Senior Research Fellow at the Reuters Institute for the Study of Journalism and leads the Institute’s research team.

Antonis Kalogeropoulos is a Lecturer in Communication and Media at the University of Liverpool and a Research Associate of the Reuters Institute for the Study of Journalism.

Felix M. Simon is a journalist, researcher, and doctoral student at the Oxford Internet Institute (OII) at the University of Oxford. He works as a research assistant at the RISJ.

Funding acknowledgements 

Factsheet published by the Reuters Institute for the Study of Journalism as part of the UK COVID-19 News and Information project funded by the Nuffield Foundation. The Nuffield Foundation is an independent charitable trust with a mission to advance social well-being. It funds research that informs social policy, primarily in Education, Welfare, and Justice. It also funds student programmes that provide opportunities for young people to develop skills in quantitative and scientific methods. The Nuffield Foundation is the founder and co-funder of the Nuffield Council on Bioethics and the Ada Lovelace Institute. The Foundation has funded this project, but the views expressed are those of the authors and not necessarily the Foundation. Visit www.nuffieldfoundation.org.

nuffield risj logo

CC BY licence logo

This report can be reproduced under the Creative Commons licence CC BY.  For more information please go to this link.