Border patrol: The rise and role of fact-checkers and their challenge to journalists’ normative boundaries

Literature: Fact-checkers and their norms

Emerging in the United States as a distinct journalistic form in the 2000s (Graves et al., 2016), fact-checkers have soared in number and visibility; 2020 figures from the Duke Reporters’ Lab (https://reporterslab.org/fact-checking/) put the global total at nearly 240 active initiatives. The phenomenon has intrigued scholars: Why is this form of journalistic activity gaining traction? All journalists check facts. What, then, is the perceived need for a separate set of people calling themselves ‘fact-checkers’ – especially when roughly a third, notably in America and Western Europe, are part of a legacy outlet with a newsroom full of journalists?

Amazeen (2020) suggests fact-checking can be understood as a democracy-building tool and a response to perceptions that democratic institutions are weak or under threat: For democracy to succeed, a society must have the means to correct citizens’ political misperceptions (Garrett et al., 2013), and fact-checking provides people with an easy way to acquire reliable civic information (Jarman, 2016). Such institutions of course include political ones; impending elections are a common impetus for the launch of fact-checking services worldwide. Yet fact-checkers also can be seen – and indeed commonly report seeing themselves (Amazeen, 2019; Graves, 2016; Singer, 2018) – as a response to concerns about another democratic institution: journalism itself. In the words of one European fact-checker, the enterprise is ‘good medicine’ for an ailing occupation (quoted in Graves, 2018: 623), one whose authority as a knowledge-certifying institution has fractured in a media environment that enables ideologues to easily convey compelling but misleading content (Jamieson, 2015).

One key to understanding fact-checkers’ place within the context of journalistic work is their approach to the fundamental norm of truth-telling. Although both fact-checkers and journalists value truthfulness, journalists tend to focus on accurately reporting what was said, while fact-checkers are more interested in judging the veracity of the statement. In taking on an adjudication role, they thus go beyond objectivity as traditionally understood in order to explicitly and overtly weigh evidence, claims and counter-claims (Coddington et al., 2014; Graves, 2016). ‘I see fact-checking as a move away from the “he said, she said” journalism that never takes a position on anything’, said Michael Dobbs (2012: 13), who founded the Washington Post fact-checker in 2007. ‘Reporters should be allowed to sift the evidence and reach conclusions’. FactCheck.org founder and long-time journalist Brooks Jackson agreed journalism should evolve quickly ‘away from the gatekeeper model, which is no longer even relevant, into more of an umpire/referee’ (quoted in Amazeen, 2013: 19).

A second crucial insight into fact-checkers’ role in addressing perceived shortcomings of traditional journalism is their emphasis on accountability and transparency. Most fact-checkers are signatories to the International Fact-Checking Network (IFCN) Code of Principles (https://ifcncodeofprinciples.poynter.org/) or have indicated plans to apply; the code expressly foregrounds commitments to transparency of sourcing, funding and organisation, and methodology. Published fact-checks typically offer links to sources used in reaching a judgement, and they commonly indicate the steps taken to investigate a statement or claim and to make a decision about it. As Graves (2017) points out, providing justification of choices made by the fact-checker is a contrast to conventional political reporting. Indeed, for fact-checkers, the practice of ‘showing your work’ is central to their claims to objectivity as they define it.

Fact-checkers are not without critics, however. They have been accused of selection bias and a (typically liberal) partisanship; of potentially confusing inconsistency (Lim, 2018; Marietta et al., 2015) in their assessments; and of over-simplification of multi-faceted and often ambiguous political statements (Uscinski and Butler, 2013). The most serious and sustained critique of fact-checkers is an epistemological one: that they discount ‘the value-laden nature of political discourse by trying to offer decisive factual conclusions about subjective questions of opinion or ideology’ (Graves, 2017: 519).

Nor is their impact on audiences clear. A study by Gottfried et al. (2013) showed fact-checking sites increase the accuracy of perceived candidate issue stances, while other scholars have found that tagging headlines as ‘rated false’ decreases their perceived accuracy more than indicating they are merely ‘disputed’ (Clayton et al., 2019). Yet considerable work has indicated attitudes and misperceptions can and do persist even when a statement is effectively discredited (Nyhan and Reifler, 2010; Thorson, 2016). Jarman (2016), for instance, has shown that motivated reasoning, particularly partisan affiliation, significantly affects the evaluation of both an initial statement and the fact-checking material around it. Indeed, a recent meta-analysis suggests that although citizens do pay attention to facts and exposure to fact-checking can have a positive influence, existing beliefs, ideology and knowledge substantially affect responses to fact-checking initiatives (Walter et al., 2019).

Nor, it seems, do many journalists know quite what to make of their fact-checking cousins (Brandtzaeg et al., 2018), or even how to define what constitutes fact-checking at all (Graves et al., 2016). But perhaps that ambivalence stems at least in part from challenges to notions of where journalism itself starts and ends.

Literature: Journalistic boundaries

Despite the inherent porousness of journalistic work, and its peculiarities as both a cultural practice and a form of knowledge production (Carlson, 2015), the openness of digital and social media has sparked numerous attempts – by scholars and practitioners – to delineate which forms of content and practice can be considered journalism and which cannot (Carlson, 2016). Of most direct relevance to this consideration of fact-checkers have been explorations of the ways in which journalists use normative concepts as definitional devices (Singer, 2015). Emphasis has gone to perceptions of the extent to which accuracy, veracity or facticity are foregrounded (Eldridge, 2016; Hermida, 2015), as well as the attention paid to independence or autonomy (Örnebring, 2013) and objectivity (Lewis, 2012; Schudson and Anderson, 2009).

In the digital age, bloggers were among the first content producers working outside legacy newsrooms to be examined through the framework of normative boundaries, with the ‘postmodern sensibilities’ of blogs (Wall, 2005: 153) – their participatory, transparent and often opinionated nature – juxtaposed with traditional journalistic values of accuracy, objectivity and fairness (Lowrey, 2006). Blogs were seen as representing an epistemological shift away from the conventions of traditional news work, offering space for more democratic and interactive discourse (Matheson, 2004). Yet beyond emphasising interactivity, the norms of bloggers were otherwise not notably unlike those of legacy journalists. Like today’s fact-checkers as well as journalists, bloggers stressed the need for accuracy and credibility (Perlmutter and Schoen, 2007), as well as truth-telling, transparency and accountability (Cenite et al., 2009; Kuhn, 2007). The boundaries between bloggers and journalists seemed quite blurry, with arguably more similarities than differences (Haas, 2006).

In the latter 2000s, so-called citizen journalists, or producers of ‘user-generated content’, became the focus of efforts to pinpoint why they were or were not doing journalistic work. Did the crucial disparity rest with core norms such as (again) objectivity (Blaagaard, 2013)? Was the issue essentially about gatekeeping and the journalists’ ability to maintain control over information dissemination (Harrison, 2010; Lewis, 2012)? Or was connection to an employer the key factor (Tong, 2015), invoking a sense of societal duty manifested through adherence to rules created within an organisational context from which external contributors are excluded (Örnebring, 2013)? At its heart, Wahl-Jorgensen (2015) suggested, the perceived boundary seemed to be rooted in the epistemology of journalism: in its ways of, and criteria for, determining what is and is not to be accorded the status of verified fact.

Over the past decade, scholars have incorporated the concept of normative boundaries in their consideration of a range of other actors doing work that is ‘journalistic’ but not necessarily ‘journalism’, a distinction also relevant to fact-checkers. Explorations of reporting by NGO staffers, for instance, address norms of independence and fairness, the dominance of traditional news values and the mutuality of relationships among information stakeholders (Moon, 2018; Powers, 2018). Other boundary work has looked at issues related to journalists’ collaboration with programmers and other big data experts. Lewis and Usher (2016), for example, have explored the interaction of programmer values such as transparency and participation with more traditional journalistic norms. Similarly, Gynnild (2016) has shown the effect of ‘computational exploration’ on journalistic practice and thought patterns.

Most of this work takes the views of journalists as the starting point, and thus considers alternative forms as representing potential normative boundary encroachment. Venture outside the newsroom, though and the tables are turned: Non-traditional content producers often see themselves as enacting norms that, they believe, legacy journalists honour mostly in the breach. Indeed, one early framing of bloggers was as a media watchdog, notably Ken Layne’s oft-cited warning to mainstream journalists: ‘It’s 2001, and we can fact check your ass!’ (quoted in Matheson, 2004: 452). Bloggers also believed they brought stories to the public that would not otherwise be covered (Kuhn, 2007). A decade later, Kus et al. (2017) found that European citizen journalists regarded traditional media coverage as biased and were motivated by a desire to exercise counter-power in political discourse.

This study examines these boundary issues through the views of another emerging set of content producers whose work overlaps with, yet is not identical to, that of legacy and digital journalists. Although fact-checkers have not generally been met with as much overt resistance as either bloggers or citizen journalists, previous work has indicated that journalists do view political fact-checking, in particular, through a prism of their own professional norms, values, motives and status perceptions (Graves et al., 2016). This study builds on that insight by focusing on the views of fact-checkers themselves, exploring their normative constructs around societal roles and audience interactions, two aspects highlighted in journalistic boundary work. It then turns to a consideration of the challenges they perceive in relation to these constructs.

Method

This study draws on two complementary data sets: interviews in 2017 with a dozen fact-checkers around the world, followed in 2018 by an online questionnaire distributed via the International Fact-Checking Network (IFCN) mailing list.

Interviews: Potential interviewees were purposively selected from a comprehensive list of fact-checkers maintained by the Duke (University) Reporters’ Lab, headed by Politifact founder Bill Adair. Selection criteria sought to maximise diversity across geography, funding models and overall approach to fact-checking, including the use of ratings, frequency of updates and types of topics covered. Gender diversity also was sought, with five women and seven men included in the final sample of 12 editors or founders at fact-checking sites across four continents. All interviewees were offered, but waived, the right to anonymity.

Semi-structured interviews, lasting an average of 1 hour, were conducted via Skype or phone in August 2017. Ten interviews were conducted in English by the author; the other two were conducted in Portuguese by a Brazilian colleague, who then transcribed and translated them into English. Interviews were recorded with the participant’s consent and supplemented with written notes. Interview data were subjected to textual analysis, a method that explicitly foregrounds the ways in which people make sense of and communicate life experiences, which in turn are influenced by communities of culture (Hawkins, 2017). The topics of normative constructs, audience relationships and occupational challenges articulated through the research questions served as an over-arching framework for this analysis. In addition to providing insights into fact-checkers’ activities and perceptions, particularly in relation to their self-identification as journalistic disrupters, the interviews also informed construction of the questionnaire.

Questionnaire: The questionnaire was created in SurveyMonkey and distributed via a link provided to members of the IFCN email list. The accompanying message specified that only active fact-checkers should click the link and complete the questionnaire. At the time of distribution in October 2018, the list included 368 recipients. However, with help from the then-executive director of IFCN, the author eliminated more than 200 duplicates, expired addresses, academics, funders and other ‘friends of fact-checkers’. The final count was a total of 161 potential respondents, of whom 34 (21.1%) took part; although some dropped out along the way, 26 answered all the questions. Relevant Ns for each question are provided in the findings.

Both open-ended and closed-ended questions were included. Because the author assumed that most normative statements, in particular, would generate high levels of affirmation, a slider was used to enable respondents to indicate the strength of their agreement, facilitating meaningful comparison of their views. Other closed-ended questions drew on Likert scale, multiple choice and structured-response formats. Given the relatively small respondent pool as well as the use of a census rather than a sample of eligible participants, descriptive statistics were used to analyse the answers to closed-ended questions.

Open-ended questions elicited a range of input, but most important to this study were descriptions of the goals, norms and perceived value of fact-checking. These were analysed through a textual analysis that facilitated question-specific groupings. For instance, responses about actions taken in pursuit of important norms (assessed through use of a slider) were grouped into categories related to ensuring accuracy, objectivity or impartiality, transparency and proper use of sources, with an additional ‘generic’ category (‘follow professional rigor’, ‘no compromise on our principles’).

All questionnaire responses were provided anonymously. Because a listserv was used to distribute the survey link, the author did not know which individuals participated, nor the extent of overlap between interviewees and questionnaire respondents. That said, several self-identified, for instance with an offer to chat through email or by referencing unique aspects of their service. ‘We have Pinocchios and people in power fear them’, wrote a respondent who obviously was with the Washington Post’s fact checker – which famously, and uniquely, uses Pinocchio icons to indicate lies. No questionnaire identifications are reported here.

Background information about respondents and their fact-checking operation was collected in the latter part of the questionnaire, and not everyone made it that far. However, only a third of the 27 respondents who answered the question said they were affiliated with or funded primarily by an existing digital, print or broadcast media outlet. The rest indicated they were independent operations. Respondents produced an average of around 16 fact-checks per month, and all but a handful said they usually or always included ratings.

Half of the 26 fact-checkers who reached the end of the questionnaire operated in Europe; the rest were from Africa or the Middle East, Asia, the USA and Oceania. All but four were IFCN signatories, and those four either had applied or planned to apply. Fifteen respondents (57.7%) previously worked as journalists; another three had been students. Other previous jobs included managing director of a charitable organisation, university professor, parliamentary advisor and scientist. Ten respondents (38.5%) had more than 20 years of prior work experience; the average was just under 15 years. However, fewer than half (12 respondents, 46.2%) said fact-checking was their only job, and four said they were unpaid for their work.

The findings address the research questions above, giving prominence to questionnaire responses and using the interview data to provide supporting evidence or context.

Findings

This section considers the three research questions in turn, focusing first on fact-checkers’ perceptions of their normative role, then on their conceptual construction of audiences and their approaches to serving that audience, and concluding with the challenges they face.

Findings: Normative role in society

That fact-checkers identify their most crucial goal as correcting misinformation (level of importance 94 on a 1–100 scale) is hardly surprising; after all, the goal of checking information is to ascertain its veracity with an eye to flagging what, if anything, is wrong. ‘Fact-checking answers a very basic need that people feel to see what they think is true reflected in public debate’, AfricaCheck.org founder and director Peter Cunliffe-Jones said in an interview. However, as Table 1 indicates, fact-checkers gave almost as much importance to simply enabling citizens to be well-informed (level of importance 92). ‘I would give media literacy a score of 150’, one respondent wrote in an open-ended space for indicating ‘other goals’. In her interview, Cristina Tardaguila, then director of Agencia Lupa in Brazil, said the site’s mission is ‘very clear: It’s to improve the public debate with quality information’. Strengthening people’s trust in information also was seen as highly important – even somewhat more so than holding the powerful in society to account. The role of fostering discussion about issues, ideas or civic affairs was accorded considerably less importance.

OPEN IN VIEWER

Table 1.

Question: One a scale of 0 (completely unimportant) to 100 (absolutely essential), how important is it to you as a fact-checker to . . ..

Aspect of fact-checking	Composite ranking on scale of 1–100 (N = 34)
Correct misinformation	94
Enable citizens in general to be well-informed	92
Strengthen citizens’ trust in information	86
Hold the powerful in society to account	80
Address claims that are not adequately investigated by other media	77
Educate voters who will be casting ballots on candidates or issues	76
Foster discussion around particular issues or ideas that are circulating in your coverage area	68
Generate a conversation about civic affairs in general	64

The actions respondents took to achieve these goals fell into three categories. The first involved presentation of the fact-checks themselves, using clear and jargon-free writing, as well as fast and frequent publication. ‘Talk in a language that people understand’, one respondent wrote; another cited the need to ‘debunk hoaxes *fast* so correct information is out there for those who want to Google’. Interviewee Mikko Salo, founder of Faktabaari in Finland, described clarity and methodological transparency as the site’s ‘visiting card’.

Transparency, along with accuracy and objectivity, also was highlighted by questionnaire respondents, part of a second category of actions: those related to enactment of journalistic norms, discussed further below. ‘Our mission is to reduce the level of deception and confusion in US politics’, one respondent wrote. ‘We do that not only by correcting misinformation, but by building trust through transparency and objectivity’.

And a third category of activities related to levels of engagement with, and responsiveness to, fact-checking audiences, also discussed further in the next sub-section. In addition to strategies for increasing visibility, several fact-checkers mentioned educating the public so citizens can do their own fact-checking. ‘We know we cannot fact-check every significant claim, so perhaps the most important thing we seek to do is to enable citizens to assess the accuracy of information for themselves’, one respondent wrote; this goal was pursued through training programs, input into school curricula and ‘most significant . . . the example we provide in our published work’. In his interview, editor Glenn Kessler of the Washington Post highlighted both normative and audience-related actions. ‘About 50% of the fact checks we do come from reader suggestions’, he said. ‘We try to be as open and transparent as possible, and as engaged with readers as possible’.

However, asked to rate the importance of various norms, fact-checkers put transparency behind several others (see Table 2). Top of the table was accuracy, though importance levels above 90 on a 100-point scale also went to impartiality (94), accountability (93), objectivity (92) and independence (91); transparency came next with an importance rating of 88. One respondent said that because they operate in countries with a ‘broad range of media standards’, they stress accuracy in order to ‘be a best practice example’. In his interview, FactCheck.org director Eugene Kiely said his role was ‘to uphold the integrity of the website, try not to make mistakes, try to be fair, try to present information in a way that will reach both sides of a partisan divide’.

OPEN IN VIEWER

Table 2.

Question: Please think about the importance of various norms or ethical values for your fact-checking. . . . How important to you as a fact-checker is . . ..

Fact-checking norm	Composite ranking on scale of 1–100 (N = 34)
Accuracy	98
Impartiality	94
Accountability	93
Objectivity	92
Independence	91
Transparency	88
Completeness or thoroughness in the information that you provide	85
Balance	71

In contrast, the lowest-rated value was balance (71), suggesting fact-checkers do not equate balance with either impartiality or objectivity, the traditional journalistic formulations (see Graves, 2016). However, several respondents used an open-ended question to point out the overlap among these three norms. As one respondent wrote:

Taking objectivity and impartiality as twin goals, the key to attaining these is to cultivate a habit of thought whereby I genuinely strip away my preconceptions about even the most trivial of factual claims or issues. Objectivity and impartiality are best as a sincere mindset or habit of thought, rather than a superficial check supplied after the completion of research.

Although transparency ranked somewhat down the closed-ended list, 14 respondents (41.2%) expanded on the norm in their open-ended comments. ‘Transparency enables accountability’, wrote one; another described the use of links to ‘original and authoritative sources’ to back up fact-checks. Several explicitly cited transparency about revenue sources, for instance by making public a list of donors. Interviewees also highlighted the importance of disclosing their editorial and fiscal autonomy. Laura Zommer, director of Argentinian fact-checker Chequeado, said the site’s reputation had been built largely on its transparency about funding, a ‘clear difference from traditional media’. Transparency about sourcing also is key.

The belief that they served as not only a complement to but also a corrective for mainstream media came through in questionnaire responses and interviews. The strongest agreement among a set of Likert scale questions asking respondents to situate their fact-checking service within the broader media environment went to the view that their offering complemented those of other outlets, as shown in Table 3. But nearly as many agreed their service was unique and an improvement over most others in their coverage area, and 70% rated themselves more trustworthy. ‘Our stress on transparency of sources and data used, and our research for clarity and balance, is something that every journalistic outlet should have. That is not the case in our coverage area, however’, one respondent wrote. ‘So I think that simply trying every day to stick to those basic principles gives us a true advantage’.

OPEN IN VIEWER

Table 3.

Question: Our fact-checking/fact-checking service . . . (N = 34).

Statement (Listed by extent of agreement/strong agreement)	Strongly agree	Agree	Neither agree nor disagree	Disagree	Strongly disagree
Offers a service that complements the offerings of other media in our coverage area.	9 (26.5%)	20 (58.2%)	4 (11.8%)	1 (2.9%)	–
Offers an improvement over most other media in our coverage area.	11 (32.4%)	16 (47.1%)	4 (11.8%)	3 (8.8%)	–
Provides a unique service within our coverage area.	19 (55.9%)	8 (23.5%)	4 (11.8%)	3 (8.8%)	–
Is more trustworthy than most other media in our coverage area.	10 (29.4%)	14 (41.2%)	10 (29.4%)	–	–
Has a greater impact on public opinion than most other media in our coverage area.	–	5 (14.7%)	21 (61.7%)	4 (11.8%)	4 (11.8%)
Has a greater impact on activities of those in power than most other media in our coverage area.	–	5 (14.7%)	19 (55.9%)	8 (22.5%)	2 (5.9%)

Respondents used space provided for two other open-ended questions – one about competitors and another asking them to describe the ‘best thing’ about their fact-checking service – to further emphasise their perceived normative strengths relative to legacy outlets. ‘We fact-check in regions of the world where media are relatively weak and governments lie’, wrote one respondent. Citing ‘servile media . . . strong on spreading fake news and spin’ as the biggest competitors, another respondent described fact-checking content as offering the benefit of ‘credibility and relevance to the public interest’. Several also cited their ability to take more time to dig into the issues, without ‘being driven by the daily news cycle’. One respondent summarised fact-checkers’ comparative benefits this way:

Our articles are of very high quality compared to the average ones from mainstream media. That is possible thanks to sound review processes, more time at our disposal to produce our content, and attention to transparency and providing all the sources to our readers.

Even in Western democracies, fact-checkers felt they addressed distinct shortcomings of the legacy media. In Austria, Fakt Ist Fakt co-founder Peter Pramer cited the difficulty that newspapers have holding the powerful to account, as they receive one of the largest shares of government advertising revenue in Europe. And Rob Edwards, chair of The Ferret, said the site was established specifically to provide a future for sustainable investigative journalism in Scotland despite the ‘deterioration’ of legacy outlets. ‘Facts and truth are important’, he said. ‘They must always have a role. As the traditional sources of information disintegrate, as they are doing, finding new trustworthy sources of information is vital’.

Fact-checkers, then, believe they fill the normative role of providing the kind of civically important, trustworthy information that legacy news outlets should be providing but, in their view, largely are not. However, as Table 3 shows, most see themselves at a disadvantage in one crucial area: impact. While they tend to believe the quality of their work is superior, they admit other outlets have a more significant effect both on the general public and those in power.

Findings: Challenges

The findings so far reflect considerable optimism among fact-checkers about the value and effectiveness of their own enterprise, particularly relative to that of traditional media. But they also indicated a number of challenges, highlighted in Table 4. The largest of these related to financial sustainability.

OPEN IN VIEWER

Table 4.

Question: Please indicate whether you disagree or agree that each of these is a challenge for you.

Challenge (N = 27; Listed by extent of agreement/strong agreement)	Strongly agree	Agree	Neither agree nor disagree	Disagree	Strongly disagree
Lack of adequate staff to support what we want to do	2 (7.4%)	15 (55.6%)	4 (14.8%)	3 (11.1%)	3 (11.1%)
Lack of adequate funding to support what we want to do	6 (22.2%)	11 (40.7%)	5 (18.5%)	2 (7.4%)	3 (11.1%)
Uncertainty about our sustainability going forward	9 (33.3%)	6 (22.2%)	7 (25.9%)	2 (7.4%)	3 (11.1%)
The small size of the audience for our fact-checks	4 (14.8%)	9 (33.3%)	6 (22.2%)	3 (11.1%)	5 (18.5%)
Difficulty in obtaining the info we need to complete our fact-checks	4 (14.8%)	9 (33.3%)	8 (29.6%)	5 (18.5%)	1 (3.7%)
Difficulty in reaching a definitive judgement in our fact-checks	2 (7.4%)	8 (29.6%)	11 (40.7%)	5 (18.5%)	1 (3.7%)
Lack of adequate office space to support what we want to do	1 (3.7%)	7 (25.9%)	6 (22.2%)	7 (25.9%)	6 (22.2%)
Lack of adequate technology to support what we want to do	–	7 (25.9%)	8 (29.6%)	9 (33.3%)	3 (11.1%)
Insufficient independence to present information in our fact-checks the way we want	–	2 (7.4%)	1 (3.7%)	9 (33.3%)	15 (55.6%)
Insufficient independence to fact-check what we want	–	1 (3.7%)	3 (11.1%)	6 (22.2%)	17 (62.7%)

Fact-checkers not affiliated with a legacy media outlet are supported by a range of revenue streams – many of them precarious, such as user donations, or of limited guaranteed duration, such as one-off grants. In response to a set of closed-ended slider questions, financial support from grants was identified as most important: from philanthropic organisations (a rating of 49 on a 100-point scale), corporate entities such as Google (37) or government agencies (32). User donations, for instance through crowdfunding campaigns, earned a rating of 30, just ahead of angel investors (27) and educational institutions (23). Very little weight was given to advertising, which earned an importance rating of only 13, and few questionnaire respondents provided training (a rating of 19) or hosted events (just 5).

Some fact-checkers highlighted this diversity in funding sources as a strength, expressed in normative terms. Interviewee Cunliffe-Jones of AfricaCheck said the mixed funding model was very important to their reputation for non-partisanship:

If you take funding 100% from the media house you work for, and that media house is seen as being conservative or pro a particular party, you will inevitably be seen as being a part of that, even if you are scrupulously non-partisan. If no organisation is providing more than 10% of funding, then it’s clear you are not acting to anyone’s will. (A diverse revenue stream) both makes you more secure and provides genuine freedom to operate more professionally.

But if financial independence is a blessing, it’s a mixed one, as shown by responses to a Likert scale question asking about the challenges fact-checkers face. Interviewee Tai Nalon of Aos Fatos said such independence was important for credibility, yet also was the ‘biggest problem’ for fact-checkers. Few if any fact-checkers are profitable, she said: ‘It’s not a sexy agenda. You are checking facts’. Although they appreciate their freedom to select and present the fact-checks they feel are most vital, they pay the price in worries about money, staff and overall sustainability, each viewed as a challenge by more than half the respondents. ‘Fact-checking takes so much time if you do it thoroughly – (a) problem in HR and money’, one respondent wrote. A majority reported operating with a full-time content-producing staff of four or fewer. More than a quarter of the 27 fact-checkers answering the question said they worried almost constantly about sustainability; another third worried a lot.

Other key worries related to content, particularly getting something wrong, a finding in line with the normative prominence given to accuracy. Perhaps less predictable – though reflecting the importance accorded to the normative goal of increasing public trust in information, as discussed above and shown in Table 1 – was a more generic concern over growing public cynicism about truth itself, which more than half the respondents said they worried about ‘a lot’ or ‘almost constantly’. In an open-ended response, one fact-checker emphasized the need to ‘get people to trust real media and be critical about those who pretend to be real’; another highlighted ‘encouraging citizens to value facts and reality over their political affiliations’. Interviewees also expressed this concern. ‘One has to have faith that people are basically good and that they want the facts’, said the Ferret’s Edwards from Scotland:

(Fact-checking) will always be crucial in a world of competing interests, money, power, whatever. People will always want someone to assess in a fair, honest, open way whether they’re being told the truth. If not, I fear for the future of civilization.

In the end, fact-checkers universally expressed the self-affirming view that their work significantly benefitted society, motivating them to struggle on against the challenges. ‘You really only do it as a value-based thing, because you think it’s right’, one respondent wrote. ‘It is too hard to live off it, because it takes time and it’s so hard to create revenue’. Asked about their primary motivation, fact-checkers expressed a conviction that the truth mattered – and that their forte was providing it. ‘I wanted to help [audiences] to be more informed, give them the information that is real, true and checked from public sources. I wanted to refute lies and participate on a project which has a potential to affect public opinions in a right way’, wrote one. Another was motivated by ‘the impact that misinformation, and lowered levels of trust in society, have on individuals and on society’. As interviewee Kiely of FactCheck.org said:

It’s gratifying when you know you’re reaching both sides in times like this. It’s hard now, it’s harder than it’s been in some time, given the political environment. But it’s still part of our goal, our mission, what we’re trying to do: to inform, not inflame.

Discussion and conclusion

It is, of course, not surprising that fact-checkers see themselves as providing a valuable public service by countering misinformation and by educating and informing audiences, as well as rising to the challenges they face because of a belief in the importance of their work. That said, this study adds to our understanding of this rapidly growing global phenomenon by expanding beyond the unique context of the US media ecosystem, as well as by exploring multiple nuances of each of these core components of fact-checkers’ self-perceptions. It also adds to the boundary literature in journalism studies, most of which is based on the perspective of those inside legacy newsrooms, by looking at key issues through the eyes of ‘outsiders’ who are performing journalistic tasks from less traditionally bounded spaces.

Indeed, the most significant finding of this study may be the extent to which fact-checkers around the world – in Western democratic societies and elsewhere – see themselves as addressing perceived shortcomings of legacy media. Not a single questionnaire respondent disagreed even slightly with a proposition that the service they offered was more trustworthy than most other media in their coverage area, and nearly 80% felt they offered an improvement over other existing outlets. The boundary literature in journalism studies has amply documented journalists’ ardent defence of occupational turf, largely on normative grounds. Yet like the bloggers of a digital generation ago, contemporary fact-checkers feel journalists are falling short in doing the job they claim that only they can do properly.

The fact-checkers identified a variety of reasons for this disconnect in perceptions, and some of those reasons were context-specific – insufficient media independence from government in one place, for instance, or a resource-driven ‘disintegration’ of local media in another. But others cut across media ecosystems. Fact-checkers highlighted several key norms, but their open-ended comments suggested that they saw independence, transparency and accountability, and the ability to be both careful and thorough in their assessments, as important advantages over legacy media. They also emphasised their role not just in informing citizens but also in educating them, a civic function that journalists focused more on ‘objectivity’ rarely foreground.

Most admitted, however, that their impact is small. Despite the rather self-congratulatory tenor of the interview and questionnaire responses, most fact-checkers remain largely dependent on others – including mainstream journalists and users, particularly via social media – to disseminate the results of their fact-checks and raise their visibility. Although sustainable funding ranked higher on their lists of challenges and worries, the two are surely intertwined: Without larger audiences and a more clearly demonstrable impact on civic society, the majority of fact-checkers who are unaffiliated with legacy media outlets will remain vulnerable. Future research might probe further into the revenue models of these fact-checking initiatives, particularly in smaller nations that currently are home to multiple fact-checkers. Of particular value would be more extensive and intensive comparisons between commercially supported fact-checkers, including those backed by profit-seeking media outlets, and those backed by NGOs or other non-profit institutions, including universities and government granting bodies. A larger sample would permit meaningful comparisons of fact-checkers working under a range of political systems, as well. Additional useful insights might be gained from an analysis of fact-checkers’ output: what they actually produce and not just what they describe themselves as producing. Rigorous content or textual analysis of fact checks is needed, as charges of bias, in particular, deserve independent assessment at an international level.

Like any study, this one has limitations. Researchers always wish for response rates to be higher – the findings here can indicate broad perceptions but cannot be taken as representative – and questionnaire results always flag up unasked questions likely to have been informative. Nonetheless, this study offers insights into the rapidly expanding phenomenon of fact-checking that are uniquely detailed as well as unusually international in scope. It also contributes to the ongoing debate around journalistic boundary-setting by further challenging the notion that journalists are the only, or even the best, defenders of their own occupational norms.