Our podcast: Who are the most vulnerable to misinformation about the pandemic?
As the second wave hits many countries, it is a good moment to pause and reflect on the challenges the pandemic has posed for governments and news organisations around the world. How informed are our audiences about COVID-19? What do they think about our work? And which people are the most vulnerable to misinformation about the disease? These are some of the questions we explore in this episode about Communications in the coronavirus crisis: lesson for a second wave, a new report funded by the Nuffield Foundation that we published this week.
Our host is our Head of Leadership Development Federica Cherubini. Our guests are Rasmus Nielsen and Richard Fletcher, two of the authors of this report. Rasmus is a director of the Reuters Institute and Richard is senior research fellow at the Institute and the leader of our research team.
Listen to the podcast
On the report's key findings ↑
Rasmus. I think we should start with the encouraging news here, which is that our work suggests that the majority of the UK public are informed, they are cautious, and they say that they are willing to take additional preventative measures to protect themselves and their loved ones from COVID-19. And many of them are also paying attention to the news, and some of them also say that they trust the information they get about COVID-19 from news organisations.
But it’s also clear that our work identifies some very real challenges, and I think we are at quite a fragile moment in the UK in terms of the communications around the crisis. When we look ahead to the second wave and the winter ahead, perhaps this will resonate with other countries too. We find, I would say, two challenges in particular. The first one is that the rally-around-the-news that we saw early in the crisis, where news consumption surged and trust in news was relatively high is really fading away, and with that fade we see not just less news consumption, less trust in news and increase in news avoidance, but also particularly growing information inequality.
We document this around age. Older people follow the news more closely than younger people. We document it around gender. Men say they follow news more closely around COVID-19 than women do, and we find it also around classic indicators of social class, like income and level of education.
Our survey didn’t allow us to really interrogate this, but I would suggest that it’s very probable there will be similar inequality around ethnicity, for example, in the UK and other countries. Of course, all of these are examples of how structural inequalities that we know from other parts of society are also present in how people use news and navigate a crisis like this.
Then finally we find that there is a growing minority of people in the UK who are using very little news around the coronavirus crisis and also express very low levels of trust in news organisations around the information they provide on the coronavirus. And this is the group that we think are potentially more vulnerable than the population at large of being, at best, less informed about the crisis and about the virus, and at worst, to be uninformed or even misinformed about the crisis that we face and what different institutions are trying to do to address it.
On the entire project ↑
Richard. Starting in mid-April, we have done fortnightly online surveys working with YouGov, and we completed 10 of those, which takes us up to the middle of August. And importantly we did this with the same people each time, or as many as we could get to do the surveys. We asked a range of different questions, not only about how people were getting news about coronavirus, but also how they felt about it, whether they trusted it and so on. And because communications are part of the official response, we also asked people about how they thought the situation was unfolding and whether institutions were responding in a good way or in a bad way.
The three main findings from my point of view that stood out are:
- Firstly, the surprisingly high levels of news use around coronavirus, and also positive attitudes towards the news. When we first started the research in April I think it was clear at that point that people were perhaps approaching coronavirus news slightly differently to how they would approach news more generally at other times.
- Then, as Rasmus has already mentioned, there was quite a sharp decrease over the summer, not only in news use but also in people’s positive attitudes towards the news as, in a sense, coronavirus, the crisis became less acute, and I think for many people it became another story amongst many.
- Finally, one thing that we haven't really mentioned in this report but something we looked at near the beginning of the research was to do with political polarisation. So it was clear right from the beginning that politics was affecting how people understood the crisis and how they behaved during it.
On the most vulnerable ↑
Rasmus. We wanted to draw inspiration from the way in which in public health research, researchers and practitioners speak about the issue of being epidemiologically vulnerable: individuals who are not necessarily ill or even necessarily going to fall ill in a situation, but are more at risk than others of falling ill, often because of structural inequalities, poverty, lack of access to resources or institutions that they can count on.
We took this notion of the epidemiologically vulnerable and then worked off of the notion of an infodemic that has been launched by the World Health Organisation, by Doctor Tedros and others- This idea that part and parcel of the coronavirus crisis is not just the virus spreading but also a situation of information overload. Some of that information is trustworthy. Some of it is ambiguous. Some of it is misinformation.
So we wanted to know who might be more exposed, if you will, and more at risk in the infodemic, not necessarily misinformed, but potentially more vulnerable to becoming misinformed. And the way we went about practically trying to estimate the size of this group was to really start just from the the observation that we know from a lot of research more broadly that paying attention to news from independent news media with professional journalists is associated with knowing more about politics, public affairs and the like. More specifically during the coronavirus crisis, we found in our own early research a similar relationship: paying attention to news from news organisations was significantly associated with people being more informed about the virus.
So the first component of how we define and measure this group of infodemically vulnerable is whether people pay attention to the news.
The second part of it is whether they trust the news, and I think this is of separate importance in that as Richard said, there’s been the ebb in news consumption as the crisis was complemented with other stories that people were paying attention to, and you could say that people simply not paying attention to the news is not in itself a sign that they’re vulnerable to misinformation or to being less informed, because they can just start paying attention. After all, they have access to the information.
The group that we define as vulnerable are both those who are not paying attention and thus likely to be less well informed about the crisis, and also who say they don’t trust the news. So even if they wanted to find out, they would be less likely to seek out news organisations, because they don’t trust them, and they would be less likely to act on the information, even if they came across it from the news organisations.
So that’s the way that we have defined the group of the infodemically vulnerable and, in the UK, by this definition, we estimate that the group has grown from about 6% of the population early on in the crisis to about 15% come late August.
On the most vulnerable ↑
Rasmus. There are really three sides to this. One is about the news that is offered to people, and I think there are many good reasons for why we have an abundance of news that’s focused on the political process around the coronavirus crisis, very good reasons to look at that. This is an inherently political crisis in many ways and there are very real political questions in how we respond as a society. But we need to recognise that a lot of people aren't that interested in politics and can be a bit turned off by the emphasis and attention paid to politicians that they often don’t hold in high regard.
And similarly I think it’s also clear there is a case for some really important journalism that is highly specialised, if you will, sort of data journalism, very intricate examinations of the epidemiology and the science behind the response to the coronavirus crisis, and that’s all very interesting and I read that all the time myself. But I think also we need to recognise again that for a lot of people news is most interesting and appealing when it’s a useful guide to everyday life, and can help people make clear decisions about things they immediately need to contend with in their own lives, so not about the sort of what’s going to happen six months from now in country X, but more what am I going to do tomorrow to protect myself and my family?
And in that sense I think there is a case here for perhaps seeking to think about whether the balance between essentially politicians and pundits versus doctors and nurses in the coverage is right. It seems that a lot of the public are not actually finding that the news that they have access to is helping explain to them what they can do to respond to the pandemic.
And I think if one could make sure that the journalism is done that really does that gets in front of people, they might think more highly of it, and perhaps sometimes there is also a case for rebalancing further towards news that people can actually really use to navigate their daily lives.
Then I think there is secondly a point about where one might reach people, and here I think it’s clear that some of this is about the news media. In the UK we have a public service like the BBC, which has a responsibility to serve all audiences, and it’s perhaps particularly important in a situation like this, the BBC really thinks about reaching, not just people like me who are well served by many different elite media that make a business off of serving my needs, but also people who are not paying for news, who are not going to upmarket and elite news sources.
And similarly of course in the private sector I think it’s particularly important to think about the role that popular newspapers and popular websites, whether legacy titles or new entrants, can play in really making sure that a wider public than just the upmarket, more affluent elite is informed about the crisis, and find information that appeals to them on their terms.
Some of this might also be about the platforms. We find that the infodemically vulnerable, for example, use social media quite a lot. Perhaps this is a channel through which news and information can reach people in ways that they find more engaging or that they come across incidentally while doing other things on the platforms.
We need to be very careful and recognise that while the platforms have been very widely used throughout the crisis, people express very low levels of trust in the information that they find on social media about COVID-19. So the platforms may be a way to get the information in front of people, but it’s really important to remember that most people are highly, highly sceptical of the information that they encounter in these environments, so perhaps an opportunity to reach them, but also, I think, some complexity in terms of whether one could actually get through and whether people will give any credence to things that they find in these environments.
On what people know about COVID-19 ↑
Richard. One of the key things we wanted to understand right from the beginning of the project was simply how much people knew about coronavirus and the epidemic more broadly. And in order to measure this we took an approach from political science, which looks at measuring political knowledge, and this is simply done by asking people a series of multiple choice questions, and then looking at how many answers people get right.
Now this was a little more challenging with coronavirus, particularly at the beginning of the project. Hard facts about the virus were still emerging, but we thought we could take information from authoritative sources in the UK and globally, and come up with a series of eight multiple choice questions, and that’s what we did.
So we tried to focus on questions that were really about what people need to know in order to stay safe from the virus, but also some questions about the broader context and current affairs surrounding coronavirus. So, to give you a couple of examples, we asked people for how long the NHS is advising people to wash their hands, the correct answer being 20 seconds, and we also asked people whether they could identify Sweden as the country in Europe that’s had the least strict lockdown.
And I think when we looked at the results, perhaps the easiest way of summing them up is that most people get most of these questions right. So over half of people get five or more questions right and that’s something we consistently found throughout the project.
Now, looking at that on its own and saying that people know a lot about coronavirus is perhaps not straightforward, but I think one way of thinking about it is when people are designing these kinds of questions to tap people’s political knowledge, it can often be difficult to come up with a list of questions that people have a realistic chance of actually getting right.
And I think it was almost the opposite situation when we were designing questions around coronavirus. It was difficult to come up with questions, as it turned out, that people could reasonably be expected to know the right answers to and get them right.
So I think we interpreted this as quite high levels of knowledge, and I think on top of this, and this is something we described in the report, is that it was clear that people tended to give quite cautious answers to questions that, although they might be incorrect, would be unlikely to leave them worse off.
So, in the example about how long people should wash their hands for, the proportion of people who selected an option that was greater than 20 seconds was larger than the proportion of people who selected an option that was lower than 20 seconds. So, even when people were wrong, they were erring on the side of caution.
On information inequality ↑
Richard. Our evidence seems to suggest that information inequality has grown over the pandemic, and what we really mean by that is we’ve seen sizable differences between groups in terms of how much or how frequently they access news about coronavirus.
If we take age as an example, at the beginning of the project in mid-April we saw that 86% of those aged 55 and over were accessing coronavirus news at least once a day, and compare that to 74% of under 55s who said the same, so that was a gap of 12 percentage points. By the time we got to August and the 10th wave of the survey, that gap had doubled to 23 points as the younger age group had stopped consuming coronavirus news at a faster rate.
That’s one example. Another is to do with gender. So again in April, when we looked at the differences in coronavirus news consumption between men and women, we saw no gap at all in April, but by August a small gap had emerged. Not a gap the same size as with age, but a significant gap nonetheless. If we look at something like education and compare the proportion of people with a university degree and those without a university degree, and look at their news consumption, we saw a gap at the beginning of the crisis that stayed roughly the, about a 10 percentage point gap that existed throughout the coronavirus crisis, so it didn’t grow any larger but it didn’t get any smaller either.
Many of these are long standing inequalities in news use. There have always been certain groups that consume more news than others, and these patterns largely reflect that. But I think we might worry about it particularly during coronavirus because the most effective response would be the one where everyone knows what they need to know in order to stay safe. And we know from other research that news has a key part to play in informing people about how they behave. So I think inequalities in news use are a cause for particular concern during a situation like coronavirus.
On lessons for the second wave ↑
Rasmus. We hope that some cautious lessons can be extrapolated from this work that’s relevant for other people elsewhere. Some of these things are aligned with what we found in comparative research we did earlier in the crisis, where we looked at six countries, including Argentina, South Korea, Spain, Germany, the US, as well as the UK. Then we found some of these things too, sort of relatively high levels of information, the public reliant on news, positive association between relying on news and being informed.
So some of these things I think have parallels with what we are expecting to see elsewhere, and that also means, of course, that some of the problems we’ve identified here or challenges in the UK around information inequality, and this reality of some people being potentially more vulnerable to being at best, less informed, and at worst, ill-informed or misinformed even, are also very relevant elsewhere.
Now, it’s also clear of course that there are some things which we should expect to be different, and I think one useful way to think about that is the point that Richard made earlier about the importance of thinking about the intersection between politics and the media in a situation like this, and that gives us a handle on some things we might expect to be more similar or more different relative to the UK.
So, for example, in our research we found that there was an early sort of high level of trust in the UK government, followed by a sharp decline, and then stability at a lower level in terms of trust in COVID information in the UK government. More broadly, polling suggests that confidence in, for example, the French government has evolved relatively similar to trust in the UK government. So, higher trust early on, sharp decline and then stability at a lower level.
But that’s not a universal pattern. There are other countries like Germany or, for example, my native Denmark, where trust has been high and very stable throughout most of the crisis according to the polling done by YouGov, and similarly of course there are other countries too, like the US and Spain, where there was low trust at the outset and there’s been stability at a lower level.
So I think this gives us some indication to some of the differences we might expect to see around the world, often tied in, in part, with this question of how polarising the coronavirus crisis has become in terms of politics, and how explicit disagreement has become around this, and how much people read the response of their government, but also other institutions like the news media through politics.
The UK and France have seen initial rallies and then high polarisation. Germany and Denmark, much lower levels of polarisation throughout and much higher levels of trust, and then countries like the US and Spain have been polarised throughout, from the outset, and lower trust throughout, and I think this gives us some indications of how things may work in the media as well, with variations from country to country.
On what reporters can do ↑
Rasmus. I want to say first that this has been a tremendous challenge for journalists as it has for citizens and societies worldwide, and I have tremendous respect for reporters who have tried to make sense of an unprecedented crisis under really, really challenging circumstances, and I don’t think there is anything in our research that could reasonably give anyone reason to blame individual journalists and the work that they’re doing, or for the fact that not all of the public is as impressed with their work as I often am.
That said, I think it’s clear that there are things that editors and journalists can think about, whether they want to do it differently if they want to make sure that not only people like me, who are news lovers, high level of education, who spend lots of time following news and lots of different sources, but also the wider public are well informed about the crisis, and equipped to really protect themselves and their loved ones, and to evaluate and judge the responses of authorities and governments and hold them to account for how they respond to the crisis.
And that I would say is to really think about whether the balance is right between elite journalism that sort of follows the intricacies of positions and the positions taken by different high level elected officials, as well as various spats between pundits who interpret hypothetical scenarios very differently.
All of this can be very fascinating and surely finds an audience, some of that audience is also willing to pay for news, but at the end of the day there is a wider public out there who are not going to read all these long reads on politics and on pundits, and who are more interested in what they should do now and tomorrow to protect themselves, and whether there’s real evidence that their government has fallen significantly short of its responsibilities in terms of protecting the public in this situation, and if so, why?
And if we want to ensure that that information is provided to people, I think we really need to think about making this as accessible as possible and as focused as possible on everyday issues that people confront in their own lives, and perhaps giving more prominence to those sources that we know from our research have not suffered the decline of trust that the news media and, in particular, the government have suffered in the UK, namely scientists and health authorities like the NHS.
So, perhaps fewer politicians and pundits, and more doctors and nurses might be a simple recommendation