March 29, 2024

Audience Perspectives on Low Trust in the Media



Photo by Jon S, Creative Commons


  

By Dr. Richard Fletcher (left) and Nic Newman (right)
Fletcher: Research Fellow
Newman: Research Associate
Reuters Institute
University of Oxford


Executive Summary and Background

This report explores the underlying reasons for low trust in the news media and social media across nine countries (United States (US), UK, Ireland, Spain, Germany, Denmark, Australia, France, and Greece).

The study is based on analysing thousands of open-ended responses from the 2017 Reuters Institute Digital News Report (Newman et al., 2017), where we asked respondents to give their reasons for low trust in their own words, using open-ended text fields. By coding and analysing responses, we categorise the specific issues that are driving public concern across countries.

Based on these categorisations, we make suggestions about ways in which journalists, platforms, and regulators could contribute to an improvement of trust over time in both the news media and social media.

Overall, we find:

Among those who do not trust the news media, the main reasons (67%) relate to bias, spin, and agendas. Simply put, a significant proportion of the public feels that powerful people are using the media to push their own political or economic interests, rather than represent ordinary readers or viewers. These feelings are most strongly held by those who are young and by those that earn the least.

  • In many countries, particularly the US and UK, some media outlets are seen as taking sides, encouraging an increasingly polarised set of opinions. Others are criticised for not calling out lies, keeping information back, or creating a false equivalence of partisan opinions that are obscuring facts and understanding.
  • In talking about trust, people mention television brands more than any other type of media (e.g. print or online). TV is considered less open to manipulation than online media, because live pictures and reporters on the spot give consumers confidence that what they are seeing is true. But TV brands are also criticised in many countries for putting speed ahead of accuracy, favouring opinion over facts, and for pushing partisan agendas.
  • For those that do trust the news media (40% across the nine markets surveyed), a significant proportion feel journalists do a good job in checking sources, verifying facts, and providing evidence to back up claims. There is more confidence in the professional integrity of journalists (and the transparency of their processes) in the US, Germany, and Denmark than in the United Kingdom, France, and Australia.
  • Social media (24%) is trusted less than the news media in its ability to separate fact from fiction. There is a sense from respondents that feeds are becoming polluted with inaccurate information, extreme agendas, and strong opinions, perhaps encouraged by social media algorithms. But, people also blame other social media users for fuelling these stories by sharing without reading them.
  • Despite this, we also find a substantial minority who trust social media for its broad range of views and authenticity. Some of these are people who distrust the mainstream media or complain about its biases and agendas. Others revel in a wide range of sources and feel confident in their ability to spot inaccurate or agenda-filled news.

Figure 1: Main reasons behind different attitudes to the news media and social media – all countries / Q6_2017_1/2. Please indicate your level of agreement with the following statements – The news media/social media does a good job in helping me distinguish fact from fiction. Base: ~2000 in each country. (click image to enlarge)

Based on these findings (summarised in Fig 1), we argue that there are a number of steps that could rebuild trust in the news, whether it comes directly from news organisations or is distributed via social platforms.

First, the news media needs to differentiate itself more from information that has not gone through the same professional checking processes. This means increasing the quality of news and minimising the clickbait that can end up deceiving and annoying consumers. Better communication of the processes that professional journalists go through to check and verify stories will also help, though not necessarily with everyone.

Second, the media need to do a better job in separating facts from opinion. Partisanship does not have to be a problem in itself, but it is widely disliked when it is dressed up as a news article or consistently spun in a way that distorts the truth. Specific media outlets – as well as journalists – need to be far more open about their biases and clearer about distinguishing news from comment.

Third, a more representative media – in terms of age, politics, economic outlook, and gender – is likely to help answer the criticism that media is only looking after the interests of the establishment. This is partly about introducing greater diversity in newsrooms, but also about plurality of provision across the political spectrum. In turn, this relates to the growing challenge of funding quality journalism commercially and the potential need for regulators to intervene in the event of market failure.

From a social media perspective, these findings pose a real dilemma. People like the range of voices and the authenticity that comes with a platform designed to encourage free expression. But this is now putting off a significant minority who worry about an increase in noise, disruptive agendas, and lack of checks. Any attempt to clamp down on misinformation is likely to restrict the breadth and vibrancy of debate. Ultimately, however, social platforms may have to find a better balance than exists today if they are not to damage their own businesses. Working with publishers, fact-checkers, and other content creators to better label different kinds of content will be important. Platforms should also consider taking into account more signals about the quality and origin of content from publishers, improving the branding of trusted brands, and taking steps to reduce the speed with which extreme or disputed content can be spread through the network.

None of these measures are likely to be enough on their own, nor will they convince everyone who currently lacks trust in the media. Restoring trust will take time and commitment. It will also need publishers, platforms, regulators, and audiences to work together if significant progress is to be made.

Background

There is a great deal of survey evidence documenting falling trust in journalism over more than a decade. In the US, Gallup has shown media trust dropping from half (53%) in 1997 to less than a third (32%) in 2016 (Swift, 2016). More widely, the Edelman Trust Barometer has documented falling trust across the world, with a growing gap between richer groups and the general population.1 The decline is not uniform (Norris 2011), but it is pervasive enough that significant numbers of citizens, even in otherwise high-trust countries, distrust journalists and news media.

Reduced trust in journalism, whether found in mainstream or social media, matters because of its role in supporting the democratic process and informing citizens so they can make choices at elections and referendums, but also in holding the rich and powerful to account. But concern has mounted in recent years about various types of false news created by Macedonian teenagers to make money and Russian propaganda farms to undermine the US elections. We have also witnessed the weaponisation of the term ‘fake news’ by Donald Trump and other politicians to describe media reports they do not like.

A recent Reuters Institute Fact Sheet on Fake News (Nielsen and Graves, 2017), based on focus group evidence in four countries, provided a bottom-up picture of how audiences define the problem. As the authors point out, the term ‘fake news’ has helped audiences express longstanding frustrations with the media environment in general – and that includes many aspects of the way traditional media ply their trade.

Figure 2: Audience perspectives on fake news (click image to enlarge)

The central point emerging from these groups was that most people do not operate with a categorical distinction between ‘fake news’ and ‘real news’, but see the difference as one of degree (see Fig 2). This study explores in more detail the continuum along which audiences see these issues. How do they view the different types of poor journalism, propaganda, and spin? How does ownership and commercial interest affect perceptions of media? What are the positive attributes of the news media that could be accentuated and strengthened?

Methodology and Approach

Our data comes from the 2017 Reuters Institute Digital News Report survey (Newman et al., 2017). This is an annual online survey of news users in 36 markets. Polling was conducted by YouGov, with
around 2,000 people surveyed in each country. Samples were based on interlocking quotas for age, gender, and region, and weighted to targets based on census/industry accepted data to better represent the population of each country.

Previous reports showed low levels of trust in the news, with less than half in many countries agreeing that they can trust most news most of the time. But we also find wide variations between countries, with trust tending to be strongest in Scandinavian countries but much lower in Greece, France, and the US.

In 2017, we wanted to explore some of the reasons for this variation, so we asked a series of follow-up questions in nine countries. The countries were chosen to represent the range of countries in our wider survey: Denmark, Germany, Ireland, and Spain at the higher end of trust; the US, France, and Greece representing countries with lower scores; and Australia and the UK somewhere in between.

In these countries, we first asked some questions to establish people’s attitudes towards different sources of information. Respondents were asked to indicate their level of agreement with the following statement: ‘the news media does a good job in helping me distinguish fact from fiction’. Available responses were on a five-point Likert scale ranging from ‘strongly agree’ to ‘strongly disagree’, with ‘neither agree nor disagree’ as the middle option. We then repeated the process with the same question, but with ‘social media’ in place of ‘the news media’.

In Figure 3, we can see the top-level results for each question (the proportion that selected either ‘strongly agree’ or ‘tend to agree’). People see the news media as doing a better job than social media in all countries, except Greece – this has more to do with the low opinion of the news media in general (just 19%) than the quality of information in news feeds (28%). Elsewhere, confidence in the news media’s ability to separate fact from fiction ranges from just 33% in France to 47% in the Republic of Ireland.

Figure 3: Proportion that agree the news media / social media does a good job in helping distinguish fact from fiction – by country / Q6_2017_1/2. Please indicate your level of agreement with the following statements – The news media/social media does a good job in helping me distinguish fact from fiction. Base: ~2000 in each country. (click image to enlarge)

Those that indicated a preference (i.e. did not select the middle option) were shown an openended text box and asked to give a reason for their choice. This created a total of 11,942 response opportunities. As is to be expected with this type of question format, and due to the difficult topic, many of those shown the open-ended text box selected the available ‘don’t know’ box to continue with the survey, or provided unintelligible answers. After these were removed, we were left with 7,915 responses that could be coded.2

Preliminary analysis of the open-ended responses identified a separate set of common codes (or themes) for each of the four attitudes described:

(i) agrees that the news media does a good job in helping distinguish fact from fiction

(ii) disagrees that the news media does a good job in helping distinguish fact from fiction

(iii) agrees that social media does a good job in helping distinguish fact from fiction

(iv) disagrees that social media does a good job in helping distinguish fact from fiction.

This allowed for the creation of a coding schema, which researchers then followed in each of the nine countries to code each response (some of which referred to more than one of the common codes).

Representative comments were translated into English for inclusion in this report. The resulting data was then cross-checked with other relevant Digital News Report data – such as demographics, political leaning, and media use – to derive further insights.

This is predominantly a piece of qualitative research. However, we do include charts throughout this report, typically displaying the proportion of responses that referred to a particular code (only the seven most popular for each attitude are shown). Because of the difficulties associated with translation, consistent coding across countries, and the complex nature of the topic, the figures we present here are only intended as a rough guide to which themes are most important and where. The figures in this report represent the percentage of those with each attitude who gave openended responses to the question as to why, and are not representative of the total population. Attempts to display data numerically can sometimes imply a spurious level of precision. As such, we focus on large differences and treat small differences between figures as potentially misleading, and we encourage readers to approach the data in the same fashion.

Trust and Mistrust in the News Media

As shown in Figure 4, overall, 40% of our sample agreed that the news media did a good job in separating fact from fiction, 25% disagreed, and 35% neither agreed nor disagreed. It is striking that people on low incomes and those below the age of 35 have less confidence in the news media (35% and 34%) than those over 35 (42%) or those on higher incomes (49%). This supports research from Edelman (Trust Barometer) which shows that richer, older, and more educated people are much more invested in the news media than the rest of the population, and this gap has been growing in countries across the world.

Figure 4: Proportion that think the news media does a good job in helping distinguish fact from fiction – all countries / Q6_2017_1. Please indicate your level of agreement with the following statements – The news media does a good job in helping me distinguish fact from fiction. Base: All respondents/low income/high income/under 35s/35s and over in all countries = 20,468/5031/4717/5276/15,192. (click image to enlarge)

Reasons for Trusting the News Media

Among those who agree that the news media helps them separate fact from fiction, a number of themes emerge from our analysis. Figure 5 shows that more than a third (39%) appear to have an inherent trust in the news media and in journalists (‘I just do’). Their personal experience is that the stories that they read or watch mostly turn out to be right. A second important reason relates to journalistic processes (22%), such as sourcing and fact-checking: ways in which the media are able to support assertions with evidence. In some countries, the depth and quality of journalism (14%) and the way stories were told emerged as a key builder of trust. The role of brands is important for many (12%), and there are often specific ones they feel they can rely on for accurate and reliable news.

Other explanations, which we will explore later in this section, include the trust that comes through being able to compare multiple sources (12%), and the role of regulators and members of the audience in keeping the news media honest (accountability, 4%). Finally, respondents talked about ‘seeing is believing’ (4%), which refers to the authenticity of live TV pictures in particular, which many felt could not be manipulated or spun like words on the printed page.

Figure 5: Why people agree the news media does a good in helping distinguish fact from fiction – all countries / Q6_2017_open_v1_agree. You said that you agree that the news media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses in all countries = 2252. (click image to enlarge)

I just do

For a significant proportion of respondents, trust in the news media is deeply held and has been earned over many years. The lived experience of this group is that the stories that the media tell mostly turn out to be true and fair. They feel that brands and journalists have a track record that means they can be broadly trusted. As we have already noted, this group is older, richer and better educated, so perhaps they have less reason to question the status quo.

Most news broadcasters in Australia have a good to excellent reputation in bringing factual news to the public. (Australia)

Mostly the news they give to us is true even if it is not one hundred per cent. (Spain)

While I don’t trust everything I read, I do believe that the majority of the reported stories I read are trying to find facts. (US)

As these comments demonstrate, there is widespread appreciation that news is often messy and that journalists will not always get things right. But this group generally finds the motivations of journalists and news brands to be honourable

I still have faith that there is some journalistic integrity left. (US)

This is a trusting group in general, with a high degree of confidence that the media can operate in a free and independent manner to pursue its role as a watchdog.

British news media is pretty unbiased – and has freedom and therefore not hampered by toeing the party line. (UK)

The media scene in Germany is free and independent to a large extent. (Germany)

The main exception was Greece, where only 16% mentioned these characteristics of the media, compared with an all-country average of 39%. As we will see later, there is far less confidence in media independence in Greece than in the other countries in this study.

Journalistic process and transparency

One of the strongest themes to emerge from our open-ended responses was the trust that is generated by the journalistic process and the transparency with which this is practised.

Confidence comes through the provision of evidence to back up claims. This will include quotes, interviews, expert analysis, and supporting statements, but also printed/on air explanations about the attempts journalists have made to check information or to present all sides of a story.

News stories are often corroborated by several reporters or witnesses, which give assurance to accuracy and reliability. (UK)

They speak to experts; they have procedures in place such as guidelines with regards to reliable sources; correct facts; and being unbiased. (US)

In recent years there has been more openness about the processes that journalists go through to verify information and check political statements. Fact-checking approaches have been widely adopted by mainstream media companies such as the Washington Post, Le Monde and the BBC, as well as by independent groups. The BBC has recently launched a fact-check brand (Reality Check) on television news and via its online service, and these initiatives have registered with many who already trust the news (see Fig 6).

BBC is good at providing explainers and presents news in a mostly unbiased way, and I know they fact check. (UK)

Sites like Le Monde do good verification work with Les Décodeurs [A fact-checking service](France)

Figure 6: Examples of fact-checking – UK and US (click image to enlarge)

In the US, a number of respondents said they regularly used independent fact-checking services like Snopes. Others saw fact-checking and verification as being a core part of the journalistic role.

Fact checking & presenting both sides of news stories are what real journalists do on the outlets I rely on. (US)

Fact-checking was spontaneously mentioned dozens of times by US respondents but only by a handful of people elsewhere. In the US, the term ‘fact-checking’ has become part of the language by which political stories are told. ‘Journalists will fact-check what the politicians say’, says one respondent who self-identifies on the political left. However, we find that these techniques are hardly mentioned by those on the political right, suggesting that fact-checking may not be seen as politically neutral by supporters of President Trump, instead having the effect of reinforcing political views. In highly polarised situations, greater use of political fact-checking might have the unintended consequence of decreasing confidence in the news media in general.

By contrast, credible and transparent sourcing by journalists seems to be respected across the political divide.

Newspapers require reliable sources and the reputable ones still have ethical standards. Even if the news is provided through a liberal lens, I can typically find the truth. (US)

In the US, unlike some other countries, journalism is seen as an important profession, with rigorous training and strong ethical standards. Those who trust the media have also picked up these cues.

Members of the media are highly educated professionals. They have a job to do and take it seriously. (US)

I trust the mainstream media content that I consume is developed by responsible, trained, professional journalists. (US)

This is not a universal view in the US, given the large numbers who do not trust the media, but the focus on trust in the profession and its practices is striking. We can see the same tendency in Germany:

News are written by experienced and well-trained journalists who check their articles on the basis of various sources before publishing. (Germany)

These are areas where we find significant differences between countries. The tradition of transparency, and editors who are charged with investigating reader complaints, in the American press can be set in stark contrast with the situation in the UK, where historically the tabloid press has tended to bury apologies in small print deep within the paper. A third of those trusting the news media in the US (32%) spontaneously mentioned journalistic process, compared with just 12% in the UK (see Fig 7).

Figure 7: Proportion that agree the news media does a good job in helping distinguish fact from fiction because of journalistic process – by country / Q6_2017_open_v1_agree. You said that you agree that the news media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: Denmark = 236, US = 267, Ireland = 314, Greece = 98, Spain = 384, Germany = 263, France = 159, UK = 273, Australia = 258. (click image to enlarge)

Transparency of sourcing and correction of mistakes is seen as an important driver of trust in many of these countries. The press in Denmark is regulated by a Press Council which the all newspapers take part in. It upholds standards and ensures inaccuracies lead to clearly marked corrections. Denmark’s journalistic processes come out strongly in our research, with 38% of respondents mentioning this aspect of trust (see Fig 7).

Counterintuitively, we find that a willingness to admit mistakes is a significant driver of trust. Audiences today seem to appreciate greater openness on the part of media organisations about what is known but also what is unknown – especially with fast-moving stories. This seems to make the news organisation more human but also make future news more believable, if there is confidence that mistakes will be put right quickly.

The best ones admit the level they are certain about the info they present and admit mistakes. (US)

Depth of content and quality of storytelling

If the UK scores poorly on transparency, there is much more spontaneous appreciation of the quality and depth of journalism (16%) carried by the BBC and broadsheet newspapers, in particular (see Fig 8).

Rather than being presented purely with the headlines a number of news programmes invite independent opinion from analysts to scrutinise the news in greater detail. (UK)

In-depth articles help you decide, you build up confidence on a particular source over time. (UK)

Investigations such as the recent Paradise Papers, or the MPs expenses scandal in the UK, are significant drivers of reputation for news brands, even if the stories are not read by everyone. In a smaller way, attempts to explain complex stories also seem to build trust with this group.

I like how some news media tries to show you the reasons behind the headlines. For example, Sky news and ITN show easy to understand graphics and diagrams. (UK)

By contrast, in Australia (5%), the US (6%) and Greece (3%) only a small proportion of respondents spontaneously talked about the depth of the investigations or the clarity with which stories were told. In these countries, improving the quality, depth, and rigour of news could significantly improve reputation and trust, at least with engaged and active news users.

Figure 8: Proportion that agree the news media does a good job in helping distinguish fact from fiction because of good storytelling and depth – by country / Q6_2017_open_v1_agree. You said that you agree that the news media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: France = 159, Germany = 263, Denmark = 236, UK = 273, Spain = 384, Ireland = 314, Australia = 258, US = 267, Greece = 98. (click image to enlarge)

Seeing is believing

The role of television as a key driver of trust (or mistrust) emerges as an interesting sub-theme when reviewing responses from our nine countries. Previous surveys have also shown that television news is trusted more than online news and social media (EBU, 2017). For many, being able to see moving images helps people make up their own mind on whether to believe a story

Seeing is believing, just like St Thomas! (France)

Images are worth more than anything that is written. (Spain)

Usually, the news media can offer images or video or interviews or statements that show what they are reporting or that they are from official sources. (Australia)

Consumers see video and audio as providing powerful evidence to support the facts: evidence that cannot be easily fabricated. Specifically, when relating to television there was also a sense that reports were more believable when correspondents could be seen on location (see Fig 9).

News companies such as the BBC and CNN already stress the ‘boots on the ground’ aspect as a reason why their content should be trusted more.

Figure 9: Correspondents on the ground help with authenticity

They usually have a reporter ‘on the spot’. (Australia)

They are usually broadcasting live and more credible. (US)

There are frequent interviews and frontline footage. (UK)

The trust for television news comes from the visibility of pictures (see Fig 9), but this often goes hand in hand with additional evidence in the form of interviews and expert opinions. It is this combination that has made television news such a key driver of trust for the public.

The role of particular brands

A key theme emerging from our open-ended responses was trust invested in a particular brand. Often this came with a sideswipe at brands that respondents found to be less reliable.

As long as you use pretty reliable sources such as BBC and not obviously partisan ones like the Daily Mail I think you can get enough information to make a judgement. (UK)

Some papers like The Express always seem to have headlines, which either are just not true which seem to come up on my Facebook and Twitter page so I always go to BBC News to see if it is correct or just made up. (UK)

In the UK, the BBC was mentioned 44 times among those who said they trusted the news media. But, as we’ll see later, the BBC also comes top in the section on mistrust. Partly due to its size – and to high levels of polarisation around contentious issues such as Brexit – the BBC is both most distrusted and most trusted.

I believe most facts that BBC say or publish. They have excellent researchers and presenters. They interview relevant people to discuss a particular issue. They appear to go beyond call of duty sometimes to investigate the truth of world news and reporters risk their lives to delve to find the truth. (UK)

Public broadcasters generate trust in other countries too, with DR (Denmark), ABC News (Australia), ARD/ZDF (Germany), and France Info (France) mentioned in positive terms. Television is under obligations to be fair and objective in most European countries as well as to provide distinctive news programming, and this approach appears to resonate with users.

It seems that DR news as well as TV2 are trying to be objective. (Denmark)

I trust the public media (ARD, ZDF). (Germany)

ABC News is by far the most in-depth and accurate I have seen. (Australia)

While television news is heavily regulated in Europe, it has a more partisan feel in the United States. Respondents tend to be fiercely attached to a particular brand while highly critical of those favouring an opposing political perspective.

I believe MSNBC approaches their reporting with care and concern. I do not believe a word from Fox News, Info Wars, Breitbart and their racist misogynistic ‘alternative facts’. (US, left-leaning respondent)

CNN and MSNBC are fair and balanced. Especially Rachel Maddow, Lawrence O’Donnell and Chris Hayes. I can’t say the same for other media news sites i.e. Fox News. (US)

Data from the Digital News Report 2017 shows that while 38% trust the news in general in the US, 53% trust the sources they use. This indicates a high degree of partisanship, much of which is generated by television brands. People seem to have stronger views about TV news than other sources, even if they use online news brands as much or more.

Newspaper brands mentioned included the New York Times in the US, Le Monde and Liberation in France, and der Spiegel in Germany. As we have already noted, those who trust the news in general tend to be richer and older, and they have a greater interest in news, so it is not surprising that broadsheet brands are praised, while partisan and tabloid outlets like Bild in Germany or Ekstra Bladet in Denmark are criticised.

Trust comes from multiple sources

For a significant minority of respondents, trust no longer manifests itself in a single anchor brand, but comes instead from the process of comparing sources.

This is particularly the case for those who are very interested in news. The internet allows different versions of events to be compared quickly, and inconsistencies can be spotted.

By getting various reports I can see different viewpoints so a more round picture emerges. (UK)

There may be times one company will provide false news, but then another company will correct the one with the false news and provide a more accurate report. It can be messy, but even the less credible news sources sometimes get things right. (US)

This type of approach is more demanding in terms of time for a consumer, and some seem to resent it. Others prefer to be in control and feel that the choice available creates more overall trust.

Accountability and the risk of publishing false information

One final issue about trust in the media relates to the reputational risk of publishing wrong information. This is partly about the risk of defamation or breaking other regulatory rules around fair access, but mostly it is the court of public opinion that respondents believe keeps journalists honest.

There is so much competition that news providers need to get it right every time or risk losing
subscribers. (Australia)

Public reaction to scandal can be very costly. Both financial and brand damnation. (Australia)

The process of reputational damage is particularly acute with brands that have the most to lose. Public broadcasters and broadsheet newspapers are trusted more partly for this reason

There are particular news media that are tasked with providing credible, reliable information. If it was ever found that they provided false information, they wouldn’t remain in business. (US)

Conclusion

Overall, a significant proportion of news users say they trust the news media to separate fact from fiction. This group tends to be wealthier, older, and more interested in news than the general population. The majority of this group are invested in the news media and loyal to one or more brands, and our research has identified a number of key reasons for this.

Mostly, this group thinks that journalists, at least from their favourite publication, are acting honourably and that they investigate stories thoroughly and professionally. Techniques to provide evidence – sources, quotes, expert statements – help to reinforce trust, perhaps to the extent that the brand or journalist becomes implicitly trusted over time. Visual clues that help show that journalists have done their homework (being in a specific location) also help build trust, as do new techniques such as fact-checking. Our research suggests that news organisations could benefit from emphasising these processes more, but mainly it is the lived experience of stories turning out to be accurate and fair again and again – over time – that is the biggest reason for trust.

This is why so much trust remains invested in traditional brands, and it may also be one reason why older groups have a higher level of trust than younger ones.

Reasons for Not Trusting the News Media

If trust is slow to build, it can also be quickly undermined. The key to improving trust may be to understand and address the concerns of those (25%) who disagree or strongly disagree with the notion that the news media does a good job of separating fact from fiction (see Fig 4). This is a minority of our sample, but it is a fair supposition that many who say they do not have a clear position on this – 35% of all respondents – share at least some of the concerns that we outline below. Younger groups and those on low incomes are much less likely to trust the news media.

Using the same methodology of studying open-ended responses, we can identify a number of key reasons for lack of trust in the news media. As shown in Figure 10, for this group, by far the biggest single reason for lack of trust is bias. Essentially, people distrust the news media because they think it has a political (24%) or commercial agenda (10%), a combination of these, or another type of bias (33%). This is often expressed in our coding as generic, because there is not enough detail to be sure about a political dimension. For many, media ownership and politics are mixed together in a way that suggests the influence of powerful interests.

On aggregate, two thirds (67%) of those who distrust the media complained of some kind of bias. The other key codes mainly relate to forms of poor journalism, such as exaggerated and sensationalised stories (11%), inaccuracy and low standards (9%), and conflicting information (3%).

Figure 10: Why people disagree the news media does a good in helping distinguish fact from fiction – all countries / Q6_2017_open_v1_disagree. You said that you disagree that the news media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses in all countries = 1657. (click image to enlarge)

Political bias

When we aggregate all biases, we see that concern about bias in the media is strongest in the US (77%) and the UK (74%), and lowest in Denmark (57%), Australia (56%), and Germany (51%), as shown in Figure 11. Concern about political bias in the media is also greatest in the US and the UK, although the true numbers are likely to be much higher than represented in the light blue bars, as we only code comments as political if they carried an explicit mention.

Figure 11: Proportion that disagree the news media does a good job in helping distinguish fact from fiction because of different types of bias – all countries / Q6_2017_open_v1_disagree. You said that you disagree that the news media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: US = 349, UK = 183, France = 159, Spain = 164, Greece = 292, Ireland = 116, Denmark = 107, Australia = 174, Germany = 113. (click to enlarge)

        United States

The reasons for high levels of concern about political bias in the US are laid bare in Figure 12, an audience map for top online news brands taken from the 2017 Digital News Report. This picture emerged by taking information about the political leaning of respondents and combining it with brand usage to show how most popular online media brands super-serve the liberal half of the US public. The further the brands are from the mid-point, the more skewed their audience is towards left/right leaning people. The size of the bubbles represents their weekly reach.

Figure 12: Audience map for the top online news brands (including new polarised news brands) – US / Q1F. Some people talk about ‘left’, ‘right’ and ‘centre’ to describe parties and politicians. With this in mind, where would you place yourself on the following scale? Q5b. Which of the following brands have you used to access news ONLINE in the last week? Please select all that apply. Base: Total sample: US = 2269. Note: Respondents who answered don’t know to Q1F were excluded. (click image to enlarge)

We find that those on the political right are almost three times more likely to distrust the news media, and that statistic is reinforced by comments from our open-ended responses, which also show the majority of all those complaining about political bias to be on the right (63%), and just 13% on the left.

Liberal media is full of bullshit and lies. (US)

Fox News keeps it fair; CNN tells us left-wing lies. (US)

They are so far to the left, they might fall off. (US)

Those complaining about political bias were almost twice as likely to be on low incomes, and were more likely to be over 50 and male.

A key theme among this group is the sense that the so-called ‘liberal media’ are wilfully misrepresenting Donald Trump’s policies and ideas. Key targets are CNN and MSNBC, but also the three big networks and newspapers such as the New York Times and the Washington Post.

I like what Mike Huckabee said, ‘If Donald Trump walked across the Hudson River, the news would read DONALD TRUMP CAN’T SWIM’. (US)

Almost everyday they present fake or deliberately distorted information. (US)

MSM showed a weak turn out for the inauguration when in fact they appeared to show crowds before the activity not during the actual Inauguration, as per CNN. (US)

Figure 13: Inauguration pictures posted by liberal media were seen as wilful misrepresentation by many Trump supporters

        United Kingdom

Our audience map for the UK (see Fig 14) appears to show that the UK media system is more balanced than that of the US, with a number of brands including the Guardian, the Huffington Post and the Independent representing the left and the Daily Mail, the Telegraph, the Sun, and the Times on the right. The majority of hyper-partisan brands in the UK, including the Canary, also represent the left. BBC News continues to be used by 70% of the UK population across radio, TV, and online, providing a common set of facts for most.

Figure 14: Audience map for the top online news brands (including new polarised news brands) – UK / Q1F. Some people talk about ‘left’, ‘right’ and ‘centre’ to describe parties and politicians. With this in mind, where would you place yourself on the following scale? Q5b. Which of the following brands have you used to access news ONLINE in the last week? Please select all that apply. Base: Total sample: UK = 2112. Note: Respondents who answered don’t know to Q1F were excluded. (click image to enlarge)

Despite this greater balance, mainstream media bias is cited as a key driver of mistrust, with most complaints coming from the left (the reverse of the position in the US). Many leftist respondents who distrust the news feel that there is a conspiracy between conservative politicians and media owners who push stories that suit their agenda. Allied to this is a strong narrative that the so-called ‘right wing press’ (the Telegraph, the Daily Mail, the Sun) has undermined the Labour party and its leaders over many years.

The Daily Mail is rampantly biased and distorts the news for its own agenda. (UK)

Of the 35 newspapers published in Scotland, 34 are right-wing. (UK)

By contrast, the anger of those on the right is less directed at the openly partisan press (the Guardian, the Mirror), but more at the UK’s public broadcaster, the BBC, which many feel has a liberal and anti-Brexit bias.

Inherent bias is not such a problem where an organisation has an explicit bias, but definitely regarding news that pretends to be neutral–- BBC/Channel 4, etc. (UK)

The media in the UK has become politically correct, feminist led, politically liberalist, anti-Brexit, anti-English, immigrant biased. (UK)

In our open-ended responses we find criticism of the BBC over Brexit from both sides of the political fence. As UK citizens have become increasingly polarised over the issue, any news organisation that tries to find a middle way is liable to be criticised from both extremes.

Brexit and US presidency coverage proved beyond any doubt the utter left-wing bias of the typical UK news media. I used to trust and value the BBC News, but now it is just a mouthpiece for the liberal leftists. (UK)

Even the BBC tends to support [a] far-right winged party (UKIP) nowadays. (UK)

As we have seen earlier, this is not a universal view. A large proportion of those who trust the news continue to support the BBC, but the polarised political climate is impacting trust in Britain’s most widely used news provider.

But attempts to be even-handed also get the BBC and other public broadcasters into trouble. By presenting both sides of an issue side by side, this can give the impression of false equivalence.

Something should not be up for discussion because it gives the sense that there are two sides both equal and they are really not. People have blatantly lied, easily fact checked and yet the media does not counter it. (UK)

This has been a particular issue in the climate change debate, where the BBC has run into difficulties by repeatedly interviewing climate change deniers, even though their claims are disputed by most scientists.

        Germany

Germany’s media system is less polarised than either the US or UK, with news organisations tending to stick to a fairly similar range of news stories and approaches. Even the tabloid Bild does not stray too far from the political centre, and there are no significant hyper-partisan websites in Germany.

In this sense, the full range of public opinion is not fully represented by the media (see Fig 15).

Figure 15: Audience map for the top online news brands – Germany / Q1F. Some people talk about ‘left’, ‘right’ and ‘centre’ to describe parties and politicians. With this in mind, where would you place yourself on the following scale? Q5b. Which of the following brands have you used to access news ONLINE in the last week? Please select all that apply. Base: Total sample: Germany = 2062. Note: Respondents who answered don’t know to Q1F were excluded. (click image to enlarge)

Although political bias is a strong theme among those distrusting the media in Germany, this is less about political partisanship. Rather, the suggestion is that the media are ‘too close’ to politicians, actively shutting down discussion of important issues such as the recent influx of immigrants

There is a lot that is hushed up in the media… (Germany)

I don’t always trust the media, especially in matters relating to the refugee crisis. (Germany)

In Germany this is also linked with a view expressed by many on the right that journalists themselves are full of (liberal) biases, which makes it even harder to write honestly about the pressures on society

The ideology of most journalists is left-winged. Politics influences journalism in a passive way, like an anticipatory obedience. (Germany)

Journalists have pre-conceived opinions and withhold information. (Germany)

Commercial bias and ownership concerns

Linked with the question of political bias is the question of newspaper ownership, commercial influence, and perceptions of hidden agendas that come with this.

This theme comes out strongly in France, which has undergone considerable media consolidation recently, and where a number of media tycoons known for their close links to high-profile politicians are perceived to be pushing their own business interests through the pages of the media they own.

The information is not independent but mostly in the hands of billionaires who dictate the editorial lines. (France)

They are overwhelmingly in the hands of oligarchs. (France)

In Greece, a number of participants talked about how the media is largely run in the interests of entrepreneurs, and that this is undermining journalistic independence and the ability of media companies to investigate political and business corruption. Fragmentation of media and the changed political situation in Greece has lessened the influence of old style tycoons, but elsewhere media remains in the hands of a few rich and influential businessmen. One case in point is Germany, where it is the concentration of ownership that worries many

The majority of the media landscape belongs to companies whose owners can be reduced to 3–5 families (Springer, Burda and Co). The one whose bread I eat, the song I sing… (Germany)

In Australia and the UK, any discussion of media ownership tends to come back to the controversial role of Rupert Murdoch, who continues to take an active interest in his newspapers and their editorial positions. His name was spontaneously mentioned dozens of times in open-ended responses among those who distrusted the media.

The bulk of printed news and to some extent that on the TV is controlled by their owners. Rupert Murdoch, is a prime example. Five print titles, and a TV network. If that’s not a stranglehold, then tell me what is. (UK)

People like Rupert Murdoch are involved in running some prominent media outlets and he has a political and social agenda. (Australia)

But it is not only ownership; others are concerned about the influence of big advertisers on the independence of media. ‘The media are subject to large corporations’, states one Spanish respondent, and these concerns are heightened by the growing financial difficulties of many big newspapers caused by declining circulation (reader revenue) and the loss of classified advertising.

They treat us like idiots, they think we do not notice that they always give the news favouring the same political or energy sector. (Spain)

Growing economic dependence on big advertisers, allied to a perception that wealthy media owners are working hand in hand with politicians to pursue mutually beneficial agendas, is affecting trust for a significant minority of respondents in our survey.

Every day we see corruption cases and nothing happens to them. All the big shots go unpunished. If the accused is a poor person, then the whole weight of the law is applied and more. All this leads us to think of the complicity of the media with the bosses. (Spain)

This may be one factor that helps explain why the young and those on low incomes, in particular, have such little confidence in the mainstream media. They are simply not seen as representing their interests against the rich and powerful.

Other biases

A range of other biases are discussed by respondents, chief among them journalistic. As we have already heard, many on the political right claim that journalists tend to be of a ‘liberal persuasion’ with a bias towards ‘political correctness’. Other complaints are more about selection of facts and their alignment to pre-existing narratives.

News only reports what they want. It is very one sided so you don’t get the full story only how they want to tell it. To report news it has to be both sides and the full story not where they want to start and finish. (Australia)

Even if large parts of a story may be factual, you can hold back on some information or use data in a certain way, to support a particular reality in a case that may have many facets. It’s not pure fabrication, but it’s not fact either. (Denmark)

In the US there was a strong sense from respondents that journalistic spin has worsened in the last few years. Much of this may relate to the increasing polarised political situation, which is reflected by polarised TV news channels such as MSNBC and Fox. But some respondents argue that this partisanship has now spread to the mainstream press:

Even the New York Times puts opinion in the news sections, masquerading as news. The more reputable media will choose content which supports their point of view, the less reputable rely on screaming headlines. (US)

News media seems to increasingly give their opinion throughout a news article. Their opinion should be saved for an editorial piece. If they are reporting news, they have an ethical responsibility to report each side of the argument. (US)

Journalistic spin is not new, but it may be that it has become more apparent. The internet allows consumers to check multiple versions of events instantly, and this often reveals inconsistencies or differences in approach.

I read a range of media and can easily see where there is evidence for factual reporting and where there is bias or the ignoring of inconvenient facts. (UK)

Depending on what channel I am watching the same story can be told on different channels… but the facts vary from channel to channel. (US)

Everyone has an agenda. The point is to know this and consume accordingly. (UK)

It is inevitable that journalistic outlets will tell stories in different ways and will choose different facts to emphasise. Journalism is a messy process, written against tight deadlines. But it seems that, without a clear explanation of these differences, many respondents are drawing their own conclusions, and these conclusions are undermining trust. The issue is mixed up in people’s minds with the motivations of the wider news organisations, suggestions of hidden agendas, and the influence of owners and advertisers. Either way, it is a toxic cocktail, which demands greater transparency and greater news literacy as a starting point for restoring trust.

Exaggeration and chasing what sells

Another reason for mistrust identified by our respondents is the belief that media companies are distorting or exaggerating the news to make money. Over the last decade, digital advertising has been the most important business model for most media companies, and this depends on generating the maximum number of clicks. In turn, this has changed the way stories are headlined, with the term clickbait entering the lexicon to describe attention-grabbing headlines. But charges of chasing what sells have also been laid at the door of television news, which has seen a ratings revival in the US on the back of partisan and sensationalist coverage of Donald Trump. ‘It may not be good for America, but it’s damn good for CBS,’ Leslie Moonves, chairman of CBS, said of the Trump phenomenon in March 2017. ‘The money’s rolling in and this is fun,’ he said. 3 Not surprising then, that people feel that it might be in the interest of media companies to overplay particular stories that are likely to drive ratings or clicks.

Seems to be all about the ratings/clicks these days. I get the feeling that news outlets churn out garbage regularly just so they can get more money from advertisers. (US)

The news tends to exaggerate, glamorise and caricature the facts (e.g. adding details), which distorts the truth in order to sell more and do better than the competition. (France)

In our open-ended responses, these issues were raised by more than one in ten of respondents (see Fig 16), ranging from 22% in Australia to just 3% in Germany.

Figure 16: Proportion that disagree the news media does a good job in helping distinguish fact from fiction because they chase what sells – by country / Q6_2017_open_v1_disagree. You said that you disagree that the news media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: Australia = 174, Denmark = 107, UK = 183, US = 349, Greece = 292, France = 159, Spain = 164, Ireland = 116, Germany = 113. (click image to enlarge)

Australian media seems to be particularly prone to clickbait approaches on the web and via television, although the public service ABC News is considered to be less guilty than other outlets.

Most forms of news, whether print, radio, internet or TV, are vultures for ‘THE’ story – and the more confronting it is, the more sales/viewers there are. They never let truth get in the way of a good story. (Australia)

I find a majority of news items are ‘sensationalised’ and often turn out to be far from the truth. They pick a headline that has very little to do with the truth of the article. (Australia)

The high score for sensationalist news in Denmark is a particular surprise. Here some respondents blame the internet and the influence of social media for encouraging even responsible brands to change their style of headlines.

Social media has encouraged clickbait, which also the big media (such as Politiken) have begun to use. (Denmark)

News media, especially on the web, is more about having a catchy heading than checking the facts. (Denmark)

Younger groups are more likely to object to clickbait and sensationalism, and women are also significantly more concerned about these practices. It is possible this is because both groups are heavier users of social media, where they tend to see more clickbait-style headlines.

Most news media tend to sensationalise everything instead of reporting the facts. I hate how they milk things for all they’re worth. (UK)

Low standards

This category contains a wide range of complaints about perceived declining journalistic standards. The greater competition associated with online media is seen as having contributed to falling standards by creating a culture where speed is now valued more than accuracy.

Many times they take news from the internet without checking them. (Spain)

Media outlets so obsessed on being the first to break a story often do this at the expense of facts. (UK)

Elsewhere, respondents take aim at a culture of celebrity journalists.

Presenters seem to be promoting themselves more than giving the facts. They interrupt don’t give the person time to answer. (UK)

In other cases, low standards are more about the ethical approach to publishing stories that editors know not to be true. The British red top tabloids, in particular, are criticised for this. Concern about low standards is often driven by seeing press coverage around a subject of personal expertise.

Newspapers in particular, are exceptionally poor, often just making things up. One only notices when they’re reporting a subject one happens to know about beforehand. (UK)

Conclusion

The reasons for lack of trust in the media are remarkably consistent across countries. Bias and agendas are rife and are perceived to have worsened with the advent of the internet. But these problems will be tough to fix, because they are driven by polarised political and economic positions, which are reflected in the media and in the personal biases of journalists themselves. After the Trump victory and the Brexit referendum, journalists have increasingly recognised the problems of representation and bias, trying to create more inclusive forms of reporting and reaching out to the other side. Some publications have also focused more on quality and investigative journalism and shifted business models to be less dependent on digital advertising. But there is little evidence that this is yet shifting the dial on trust.

Indeed, the worsening financial situation is making many news organisations even more vulnerable to the views of big advertisers or powerful owners, who are prepared to support them financially in return for influence.

Trust in Social Media

Respondents across all nine countries were also asked whether they think that social media does a good job in helping them distinguish fact from fiction. Unlike much of the news media, which often claims this as their raison d’être, most social networks do not claim that the separation of fact from fiction is a service they provide. Despite this, we have a seen a dramatic global increase in the proportion of people that say they use social media for news each week, with figures doubling from around 25% to around 50% in the last five years across the countries we survey in the Digital News Report. Clearly, then, a significant chunk of the population turn to social media to learn about what is happening in the world.

We should, however, avoid conflating ‘use’ and ‘trust’. Throughout the twentieth century, people consumed news from sources they said they had little or no trust in. Today, some of the most popular newspapers in Europe are tabloids that often find themselves at the bottom of trust rankings. This helps us understand why the use of social media for news can grow even as trust levels are relatively low. Across all countries, just 24% say that social media does a good job in helping them separate fact from fiction, compared to 40% for the news media (see Fig 17). Unlike with the news media, this figure of 24% varies little with key demographic variables, such as age, gender, education, and income. While a lack of confidence in sources of information has clear downsides, we might also be reassured by this figure, as it also suggests that people are approaching information on social media with a large degree of scepticism. This is something that is often absent from discussions about misinformation on social media. Similarly, the idea that younger people, or those with low incomes and low levels of education, are more trusting of news they come across on social media is an assumption that is not supported by our data.

Figure 17: Proportion that agree social media does a good job in helping distinguish fact from fiction – all countries / Q6_2017_2. Please indicate your level of agreement with the following statements – Social media does a good job in helping me distinguish fact from fiction. Base: All respondents/low income/high income/under 35s/35s and over in all countries = 20,468/5031/4717/5276/15,192. (click image to enlarge)

Reasons for Trusting Social Media

Among those that agree that social media helps them separate fact from fiction, two clear explanations emerge from our analysis of the open-ended responses. First, around a third (33%) say that it is because social media exposes them to a broader range of sources and views. Second, 27% say that it is because social media is more authentic, with ‘real people’ allowed to express their views (see Fig 18).

A handful of other explanations were given, but were less widespread. These, which we shall explore in more detail below, include: being able to see the views of people they know (9%); being able to select to follow only sources they trust (8%); that mistakes and misinformation are corrected quickly (6%); and that news access is convenient and up to date (5%). Unlike with the news media, the proportion of people that say they inherently ‘just trust’ social media is lower. This might suggest that people have a clearer sense of why they should or should not trust social media.

Figure 18: Why people agree social media does a good in helping distinguish fact from fiction – all countries / Q6_2017_open_v2_agree. You said that you agree that social media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses in all countries = 1197. (click image to enlarge)

Broader range of sources and views

The most commonly stated reason for believing that social media does a good job in separating fact from fiction was that it provides access to a broader range of sources and views. In other words, having access to information from lots of different sources allows a consensus to emerge, or highlights where there are disagreements or gaps in what is known.

There are enough different voices that you can piece together what actually happened. (UK)

My social media feed shows multiple news reports from a variety of trusted media sites, which either report similar/same stories that lead me to conclude the story is truthful. (Australia)

Different views appear on social media platforms and one can form an opinion about a specific topic. (Germany)

I can view multiple sources on the same issue or event, and there are people critiquing the articles and posts, so I can see where some things were skewed or reported narrowly. (US)

Some responses, like the bottom quote above, are clearly motivated by the perceived failings of the news media. It would be a mistake to assume that people with low confidence in the news media automatically have high confidence in social media – in fact, having a low opinion of one is correlated with having a low opinion of the other – but some people clearly think that social media can correct some of the news media’s shortcomings.

This emphasis on breadth also chimes with our other research that has shown that people who use social media tend to use more sources of news than those who do not (Fletcher and Nielsen, 2017). This is primarily driven by ‘incidental exposure’, where people are shown news content even as they use social media for other reasons. Even for those people that use social media specifically for news, they may still be shown news from sources they would not otherwise use, with the effect of increasing the diversity of their news diets. These processes are in part driven by algorithmic selection, which means that they are common to search engines and news aggregators.

Although it is among the most popular reasons in every country, as Figure 18 shows, breadth is more valued in Denmark (41%) and Greece (43%), than in Australia (25%) and the UK (26%).

Figure 19: Proportion that agree social media does a good job in helping distinguish fact from fiction because of the broader range of views/sources – by country / Q6_2017_open_v2_agree. You said that you agree that social media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: Greece = 174, Denmark = 65, Spain = 261, UK = 102, Ireland = 151, US = 102, Germany = 95, France = 78, Australia = 170. (click image to enlarge)

Authenticity

Related to the diversity of sources and views that social media makes available, people also value it because of its authenticity. For some, social media does a good job in separating fact from fiction because it lets them hear the views of ‘real people’. For others, the authenticity stems from the availability of unedited content that has not been subjected to journalistic processes.

I trust the common man over the left-wing orientated press. (Germany)

The real news comes from real people with no influences from financial or political gain. It also helps to fill in the gaps that mainstream media leaves out, so I get the whole picture and not the financially/politically invested news. (Ireland)

Most of the people who mentioned authenticity did so in ways that suggested that journalism contaminates descriptions of events. The idea is that journalists will take a story and distort it to suit a particular agenda. A story might be sensationalised to make it more commercial, or modified to fit with a political worldview.

Some comments were more specific than this, describing how newsworthy people can now use social media to communicate directly with the public. In this situation, social media is valued because you can get news ‘straight from the horse’s mouth’ (UK).

Some respondents used the open-ended answers to refer to specific events that they felt were not given adequate coverage by journalists, but had appeared on social media. One respondent in Ireland mentioned the coverage of controversial 2015 government proposals to charge for water use, and the belief that the public protests opposing this policy was being covered up.4 In the UK, too, the idea that news organisations are deliberately choosing not to cover certain events – often large protest marches – for political reasons has become a common refrain, even when these events have received coverage.

With social media, you get to see pictures of events that are sometimes kept from viewers on TV. Example being the water protest marches. (Ireland)

Figure 20: Proportion that agree social media does a good job in helping distinguish fact from fiction because of authenticity and the broader range of views/sources – all countries, by age / Q6_2017_open_v2_agree. You said that you agree that social media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses from 18-24s/25-34s/35-44s/45-54s/55+s: 118/208/234/235/402. (click image to enlarge)

Considering both breadth and authenticity, there are some differences when we split the data by age. Figure 20 shows that authenticity is valued slightly more among older people, with 31% of over-55s citing it, compared to under 30% for every other age group. But the most striking difference is the gap between authenticity and breadth among the 18–24s. Breadth (41%) is clearly more valued than authenticity (17%) among this age group, who are by far the most active social media users. The UK was the only country where authenticity was the most popular reason (43%) for trusting social media, ahead of breadth (see Fig 21). Here, there have been a number of high-profile press scandals that have shaken confidence in journalistic methods. Furthermore, the debates leading up to the Brexit referendum highlighted divisions between those that embrace and those that reject expert or elite opinion.

Figure 21: Proportion that agree social media does a good job in helping distinguish fact from fiction because of authenticity – by country / Q6_2017_open_v2_agree. You said that you agree that social media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: UK = 102, Germany = 95, Spain = 261, Greece = 174, Ireland = 151, France = 78, Denmark = 65, US = 102, Australia = 170. (click image to enlarge)

Following particular sources, self-correction, and convenience

Around one in ten people across all nine countries say that they think social media does a good job in separating fact from fiction because it allows them to see the views of their friends and other people they know. The idea here is simply that people trust their friends as a guide to what is important, and that social media allows them to quickly and easily see what that is.

I think you can quickly sort out what is fact or fiction through other people’s reaction or comments. (Ireland)

Because debates on social media and news that gets shared on it will show me if something is fake news. (France)

But as the quotes above highlight, this is also connected to commenting in a general sense. People believe that interactivity around information ultimately helps them to understand the veracity of stories, as others can offer feedback and suggest alternatives. There were only small national differences in the extent to which the views of friends are valued, but this reason was very slightly more common among younger people and women.

Around 6% stated that the interactivity around social media in a sense makes it self-correcting. A user might post something that contains misinformation but, in their experience, other users will jump in to correct it, leaving people with a more reliable version.

If something isn’t right or untrue someone will correct it online but if it’s on the news you just have to accept it. (UK)

If it’s false it’s shot down pretty quickly. (Australia)

In some countries, the news media are often seen as reluctant to correct things they have published in traditional formats, because the corrections are not given the same prominence as the original story. In response, screenshots of corrections issued by newspapers can go viral on Twitter, as people attempt to make them as widely known as possible. Online versions of stories are often updated to correct mistakes, but these edits often go unnoticed by readers.

This explanation was not particularly popular in any of the countries we studied, with the highest proportion in Greece and Denmark (11%). However, it was slightly more popular with young women.

Two more infrequently cited reasons were the ability to follow trusted sources (8%), and the speed and convenience that social media offers (5%). Both of these activities require some active curation from the user (changing settings, following news pages, etc.) in order to make the most of them. Given that we know from the Digital News Report that this is a minority activity (for news), it is unsurprising that these reasons were only mentioned by a small minority).

I trust information posted by my reliable contacts or by newspaper social media pages.(France)

[Social media is] too immediate to be manipulated. (France)

Time is restricted so news is brief and to the point. (Australia)

Though not as commonly cited among those that think social media does a good job in separating fact from fiction, 17% of respondents still offered explanations of this type.

My country tends to be trustworthy and the general population fairly sensible. (Australia)

They are up to date and usually real news. (Spain)

As with previous sections, some of the respondents drew on positive personal experiences to arrive at this view. For others, it may simply be that they find it difficult to articulate reasons for their trust in social media. ‘I just do’ was most commonly cited in Australia (31%) and Germany (30%).

Conclusion

We should keep in mind that just a quarter of people think that social media helps them distinguish fact from fiction. But among those that do, their reasons essentially link back to the fact that it can give voice to a broader range of actors – whether news organisations, ordinary people, or friends – and that information can be quickly added, updated and corrected. Of these, it is clear people find breadth to be the most beneficial.

It is perhaps striking that relatively few people think that one of the things that makes social media truly distinct, namely the level of complex interactivity around information, helps them separate fact from fiction. However, it is important to remember that only a minority chose to make use of the full interactive capabilities of social media, as previous Digital News Reports have shown. Rather, people appear to value, above all else, the ability to know what those they like and trust are thinking.

Reasons for Not Trusting Social Media

As we saw from Figure 17, the proportion that do not think that social media helps them separate fact from fiction (41%) is significantly greater than those that think it does (24%). Again, this suggests that many people take quite a critical approach to information they see on social media. Our open-ended responses help us to understand why this is.

Figure 22: Why people disagree social media does a good in helping distinguish fact from fiction – all  countries / Q6_2017_open_v2_disagree. You said that you disagree that social media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses in all countries = 2635. (click image to enlarge)

The three most popular reasons given, as shown in Figure 22, were that the information they typically saw on social media was unreliable or of low quality (35%), that there were no checks on the authenticity of the information (25%), and that the information was too opinionated, biased or agenda-driven (24%).

Concerns to do with clickbait and sensationalism (6%), viral information (5%), and information overload (5%) were mentioned by a handful of respondents, but not nearly to the same extent. Only 7% offered reasons along the lines of ‘I just don’t’, suggesting that many people who have low confidence in social media are able to articulate why this is.

Low quality/unreliable

Just over one third (35%) of those who do not think that social media helps them separate fact from fiction said that this was because they think that the information is low quality or unreliable.

This is a standard criticism of information on social media, and it echoes what is now a decadesold concern about online information more generally. However, it also lacks specificity, with open-ended responses coded in this way if they identified the information they see on social media as flawed in some way, without explaining why. There is a sense in which this theme overlaps with all of the others, or rather, the other themes in this section can perhaps be thought of as subcategories of this broader explanation.

Agenda-driven/opinion and lack of checks

We see more interesting variation when we examine the other responses in this category. Some referred to the idea that there is too much agenda-driven or opinionated information on social media.

You come into contact with opinions, which are always biased, so you’re only ever getting personal perspectives on events. (France)

Others picked up on the lack of checks and accountability.

Social media tend to reflect people’s often uninformed opinions. Most people do not seem to put the same amount of thought, effort or research into their posts as journalists put into their work. (Ireland)

It is very easy to post fake news on any topic, such as announcing a death, which is what happened with Arnold Schwarzenegger. (France)

These are quite different concerns, but we can see that – leaving the generic low-quality complaint aside – people tended pick one of these two as their primary reason for not believing that social media does a good job in separating fact from fiction. In Greece (34%), France (30%), Spain (25%), and the US (31%), where the news media tends to be highly polarised along political lines, the presence of highly opinionated information is a relatively widespread concern (see Fig 23). In these countries, we would expect that people would be more likely to be exposed to cross-cutting information, or information in general is more likely to be shaped by ideology.

In countries where the news media is far less polarised – such as Denmark (13%) and Germany (9%) – this is a seen as less of a problem. A more widespread concern is the lack of checks and balances around who can post information, and whether the information itself is verified. Just under a third in Denmark (30%), and just over a third in Germany (39%) identified this as the main reason why they have little confidence in social media (see Fig 23). These countries have deeply engrained norms and practices associated with professional journalism and lower levels of political polarisation, and the dissemination of information has historically been the preserve of elite individuals and institutions, although, as we have already seen in the first half of this report, some believe that arrangements like these have downsides.

Figure 23: Proportion that disagree social media does a good job in helping distinguish fact from fiction because of lack of checks and opinionated information – by country / Q6_2017_open_v2_disagree. You said that you disagree that social media does a good job in helping me distinguish fact from fiction. Why is that? Base: All intelligible responses: Greece = 221, US = 463, France = 221, Spain = 258, Australia = 277, Ireland =273, UK = 364, Denmark = 270, Germany = 288. (click image to enlarge)

Sensationalism, virality, and information overload

Clickbait, exaggeration, and sensationalism were mentioned much less frequently by respondents in all countries. The highest national figure was in Denmark, where 10% cited this issue, but in Germany and France the figure was just 3% (see Appendix). Information of this type may be frequently encountered on social media, and seen as an annoyance by many, but it is possible that most people feel that it does not stop them from being able to separate the true from the false.

Many social media news stories contain clickbait headlines and facts backed up with limited citations. Seems to be more common in ideologically biased sources, regardless of side (e.g. Buzzfeed, GetUp or Breitbart). (Australia)

Virality is a separate but closely related issue. Whereas clickbait and sensationalism are primarily the product of journalists and headline writers, virality is a consequence of the scale, speed, and responsiveness of social networks. Because social networks rely on algorithms to make rapid procedural decisions, information can go viral and reach huge numbers of people with remarkable speed. News organisations know this, and some are willing to exploit it by creating superficially popular content that will be raised to prominence due to the activity surrounding it. James Webster (2014) has called this the ‘popularity bias’, the obvious downside being that popular information might take the place of information that helps users separate fact from fiction.

Often inaccurate articles spread like wildfire before facts have been checked. (Ireland)

Social media can rarely if ever provide all the necessary facts of a situation as news is presented in ‘bite-size’ form for maximum shareability. This is easily exploited or misused to spread false or misleading information, or opinions presented as fact. I remain sceptical of social media headlines and any news that presents a clear bias. (Australia)

A much more straightforward concern voiced by 5% of respondents was simply that there was too much information on social media. Information abundance is one of the key features of the digital age, and it is not difficult to see how this might make it difficult to separate fact from fiction, particularly if much of the available information is contradictory.

Some believe that the problem is particularly acute on social media, because it ‘flattens’ information environments, so that sources that differ in important ways are given equal prominence. Within news, this has allowed new players to gain a foothold in a market that was historically tough for new entrants to compete in. This has increased information diversity in a rather crude sense, and some are deeply concerned about the potential consequences. But, at least for the time being, this is not a concern that many social media users share.

It’s not always easy to find sources… Contradictory articles and trolling and conspiracy theory websites are everywhere and sow trouble. (France)

Conclusion

Many people approach information on social media with a high degree of scepticism. The primary reasons for this all seem to relate to information quality, either because it is seen as too opinionated, unreliable, or sensationalised, or simply because there’s so much information that it becomes hard to separate the good from the bad.

It is interesting that so few respondents (5%) explain their lack of confidence in social media in terms related to virality and algorithmic selection (see Fig 22). In general, people do not readily talk about the technical properties of social networks, and instead focus on how other people use them to spread low-quality information. This may be linked to a widespread lack of ‘algorithmic awareness’ (Eslami et al., 2015), or perhaps a recognition that algorithmic selection has benefits as well as drawbacks (Bucher, 2017). Either way, it is clear that many people see other users – whether journalists or members of the public – as primarily responsible for disinformation on social networks.

Discussion and Recommendations

Our analysis of the open-ended responses reveals a wide range of reasons for low levels of trust in the news, and our interpretation highlights four important themes.

First, in many, but not all, of our countries, we see a growth of media partisanship. Instead of representing all parts of a community, news organisations increasingly take sides, run campaigns, and drive agendas. In many cases these agendas are hidden, representing the interests of powerful political or commercial interests. To some extent this has always been true, but social and digital media has made this more apparent. This is no longer just played out on the opinion pages, but has started to bleed into the news itself. Our open-ended responses show these biases and agendas represent the biggest single reason for low media trust.

Second, many people no longer see the media as representing the interests of ordinary people, particularly the young and those on low incomes. Given fast-changing and increasingly multicultural societies, there is a problem of representation in newsrooms, in terms of age, gender, class, political outlook, and ethnic background. For many people, the news media is seen as part of the establishment elite, biased, or just out of touch.

Third, the changing economic models are seen as lowering journalistic standards. Intense competition for attention is changing the type of stories that are commissioned and the way headlines are written, misleading and confusing audiences. In television news, the drive for ratings, and the drive for clicks on the internet, is viewed as rewarding sensationalist, emotive, and partisan news. Accuracy is often a casualty as newsrooms strive to be first. Audiences have noted these changes and have drawn their own conclusions.

Fourth, the growth of the internet and the rise of social media has created a world in which multiple perspectives have become the norm. The sheer abundance of information can create confusion as much as clarity. On any story, people are routinely exposed to different opinions and alternative facts, whereas previously they might have stuck to a single source. This, in turn, has led people to question the integrity of news organisations that had previously been beyond reproach. Politicians and bloggers use social media to further question the motivations of journalists and news organisations – in a further cycle of mistrust.

Given the speed and scope of these changes, it is not surprising that trust in journalism has fallen overall. But it is worth noting that not everyone is complaining. Many respondents base their express trust in news on core professional principles and practices of sourcing, verification, and commitment to finding facts and reporting them. These values are clearly still important for many news users, even when people often feel journalists and news media fall short of them. Furthermore, digital and social media have provided a vast range of new perspectives, and the ability to communicate and discuss the news across the world. Many of our respondents are deeply sceptical about the news – mistrustful even – but they are also better informed and enjoy access to a much wider range of sources. Nobody in our open-ended responses said they wanted to go back to a world with a small number of sources. Perhaps falling trust is an inevitable consequence of the increased range and availability of news and the greater opportunities to participate. Perhaps we need to consider discounting some trust in favour of scepticism and a focus on news literacy.

We pick up this idea most strongly in the sections on trust and distrust in social media, where benefits and dis-benefits so clearly reflect two sides of the same coin. People love the range of views and the authenticity in social media but complain about agenda-filled feeds and the low quality of information. Both are the direct and inevitable consequence of a model that allows anyone to publish without checks. But knowing this does not make the problem of misinformation any easier to solve. More checks, or any greater algorithmic priority given to ‘authoritative’ news sources, will narrow the range of voices. Once again, perhaps a narrow focus on trust will lead us to undervalue the wider benefits associated with digital media.

There is also much we are yet to uncover. The high number of ‘don’t know’s, and the widespread use of ‘I just do’ style responses highlight that this is a complex topic, where many find it hard to articulate specific reasons for their level of trust in different media. As long as this is the case, measures designed to restore trust may lack focus.

So how then can we square this circle? In the light of our specific findings here, what measures would enable us to increase trust at the same time as maintaining range and choice? What should media companies do and how could technical platforms like Google and Facebook help? (see also recommendations from Wardle and Derakhshan, 2017).

News Media Responses

The critical challenge for media companies is to distinguish their journalism better from the mass of information available on the internet. Our research suggests that trust could be improved by emphasising more clearly the processes that professional journalists go through to verify and check the news. This is not just about fact-checking, but the day-to-day process of being transparent about sources, standing-up stories, and being open about mistakes. Journalists can also make an impact by taking these techniques into the social media sphere and challenging misinformation there.

Equally important could be raising the quality of journalism, producing more investigative and in-depth reporting that stands out from the mass of general news. The shift of business models away from digital advertising and towards reader income could reduce the need for clicks and encourage a greater focus on quality. Not all news organisations will go down this path, but the Guardian, the New York Times and the Washington Post have already committed themselves to a future based on reader loyalty, as have digital-born news organisations like De Correspondent, El Diario, and MediaPart.

News organisations also need to address the various problems of representation and inbuilt bias that are raised by this study. ‘If journalists become distant from other people’s lives, they miss the story, and people don’t trust them,’ said Guardian editor Kath Viner in a recent speech (Viner, 2017). She has pledged to do more to address diversity in the newsroom and to reach out to include a more diverse set of perspectives. In the wake of the Grenfell Tower fire, Channel 4 presenter Jon Snow has noted that the failure to pick up warnings on safety showed that the media were ‘comfortably with the elite, with little awareness, contact, or connection with those not of the elite’.

The question of bias and hidden agendas must also be tackled. Partisan news organisations could increase trust by being clearer about their biases and funding. Others, who aspire to be trusted by broad sections of the community, will need to do more to rid news coverage of agendas and tired narratives, and bring a wider range of opinions into their coverage.

Technical Platforms

Low trust in social media may not be affecting usage, but the publicity around misinformation is affecting business and attracting the attention of regulators.

The major platforms are already engaging in two types of initiative that aim to deal with different aspects of the problem. One looks at the supply side and the other at demand.

On the supply side, platforms are trying to identify and disqualify so-called bad actors (domains or individuals). This is straightforward if platform rules are violated, but defining bad information is notoriously difficult so Facebook and Google have tended to work with third-party fact-checkers. Disputed claims are then surfaced at the appropriate point within search or social media. It is still early days, but the effectiveness of these approaches has already been questioned because of the type of confirmation bias noted earlier in this study. Political fact-checking of polarised societies may end up further entrenching people in their beliefs.

Figure 24: Fact checking initiatives on Google and Facebook (click image to enlarge)

Another approach is to identify organisations that can be trusted, and surface their news more often in algorithms or through other appropriate labelling. Verification ticks in Facebook and Twitter were a good early example of these signals, but tech platforms would also like to pay more attention to the trustworthiness of a piece of content. However, this raises the difficult question of who should decide which content or organisations are more or less trustworthy.

The non-partisan Trust Project has developed a series of trust indicators based on interviews with ordinary people and consultation with news publishers.6 These include information about publisher policies on ethics, fact-checking, and corrections, as well as their ownership structure. Facebook and Google are already starting to incorporate these signals.

As this study has shown, one of the key indicators of whether content can be trusted or not is the news brand itself. Reuters Institute research has shown that the news brand is recognised less than half the time when content is accessed from Facebook or Google (Kalogeropoulos and Newman, 2017). Providing stronger branding within platforms could improve trust levels considerably by reducing potential confusion. Google has already introduced better branding within its AMP product, and Facebook is rolling out better branding in trending news, but not yet in the news feed itself.

Working with News Consumers

On the supply side, there are initiatives to improve the digital literacy of news consumers. This study documents the considerable frustration felt by many at the way in which social media users, for example, share inaccurate or inflammatory content, without reading it or thinking about the consequences. The News Integrity Project, funded by Facebook, the Craig Newmark Foundation, and others, also has a focus on news literacy. It has pledged $14m to build trust between newsrooms and the public, and to make public conversations more fruitful and less polarised. Early funded projects include more community involvement in investigative reporting, and better comments.

Many of the initiatives outlined above will make a difference; some will be less successful, but it is in everyone’s interest to keep trying. Rebuilding trust will a long-term process and will require the commitment of publishers, platforms, and consumers over many years. Finding solutions requires a solid understanding of the perceptions and motivations of consumers, and this paper aims to contribute by providing at least a snapshot of current concerns. In many ways, we find audiences ahead of publishers and platforms in demanding change.


Originally published by Reuters Digital News Report under a Creative Commons Attribution 3.0 Unported license.