Establishing Trust in Experts During a Crisis: Expert Trustworthiness and Media Use During the COVID-19 Pandemic

Abstract

Existing research on factors informing public perceptions of expert trustworthiness was largely conducted during stable periods and in long-established Western liberal democracies. This article asks whether the same factors apply during a major health crisis and in relatively new democracies. Drawing on 120 interviews and diaries conducted during the first wave of the COVID-19 pandemic in the Czech Republic, Hungary, Poland, and Serbia, we identify two additional factors not acknowledged in existing research, namely personal contact with experts and experts’ independence from political elites. We also examine how different factors interact and show how distrust of experts can lead to exposure to online misinformation.

Introduction

The key role played by public trust in enabling the successful implementation of public health policies is by now well established. Studies have shown that a lack of trust in health experts can obstruct the implementation of public health policies, including vaccination campaigns (e.g., Larson & Heymann, 2010; Stecula et al., 2020). Trust in science, along with trust in politics, was also a reliable predictor of the adoption of protective measures during the COVID-19 pandemic (Dohle et al., 2020). At the same time, it is important to acknowledge that trust can easily become harmful if people misjudge the trustworthiness of individuals or institutions and rely on misleading information and advice (Hosking, 2014; O’Neill, 2002). The risk of being swayed by untrustworthy information and advice is particularly prominent in the contemporary, hypermediated environment, marked by the growing dominance of digital media. This environment eased the diffusion of expert knowledge and enabled greater public engagement with science, but also brought new challenges in the form of misinformation and public controversies that can undermine trust in expertise (Davies & Hara, 2017; Van Dijck & Alinejad, 2020). Arguably, these challenges have a negative effect not only on public trust in experts as such but also on people’s ability to identify trustworthy expert information.

Despite the importance of public trust in experts, and recent changes in the information environment that limit people’s ability to identify trustworthy information, research examining the factors that shape people’s perceptions of expert trustworthiness remains relatively limited. Existing work identifies several indicators that inform lay people’s judgments of expert trustworthiness, including perceptions of competence, adherence to scientific standards, and good intentions (Besley et al., 2021; Hendriks et al., 2015). Studies have also shown that perceptions of trustworthiness can be affected by the presence or absence of expert consensus on the topic, and by expert’s choice of language and channel of communication (e.g. Gustafson & Rice, 2019; König & Jucks, 2019). However, this body of work was conducted largely outside of a crisis context and in long-established liberal democracies of the West. We therefore have little understanding of how people’s judgments of experts and their trustworthiness might change during a major public health crisis such as the COVID-19 pandemic, or how they might operate in different sociopolitical environments.

Available knowledge on the impact of the changing information environment on public trust in experts is limited as well. While existing research provides important insights into the relationship between media use and trust in experts, as well as related phenomena such as trust in science (e.g., Anderson et al., 2012; Dudo et al., 2011; Nisbet et al., 2002), it typically assumes a one-way relationship between media use and trust, whereby the media are treated as sources of information that shape trust. Such an approach fails to capture the reality of communication in a high-choice media environment, in which citizens are faced with an increasingly complex and diverse information environment (e.g. Van Dijck & Alinejad, 2020) and in which trust in experts can itself be a determinant of media use. It is therefore pertinent to ask how the relationship between public trust in experts and media use might change in the context of a communication environment characterized by a high choice of information sources and proliferation of misinformation.

To tackle these gaps, this article examines trust formation and media use during the first wave of the COVID-19 pandemic, drawing on 120 qualitative interviews and diaries conducted in four east European countries: the Czech Republic, Hungary, Poland, and Serbia. Several factors make a focus on Eastern Europe particularly appealing. The region has one of the lowest levels of trust in science, as well as the lowest levels of support for vaccination globally (Sturgis et al., 2021). It is also characterized by low levels of political trust (Závecz, 2017) and has been affected by a rise in illiberal tendencies and populism in recent years (Vachudova, 2020)—all factors that have been associated with lower levels of trust in science.

Literature Review

Trust in Experts

Trust is a notoriously elusive concept, which is both difficult to define and measure. It is often conceptualized as a property of individuals—typically, as an individual-level belief, attitude, or even an aspect of personality, which involves the perception of others as reliable, and an acceptance of one’s vulnerability to the actions of others. For instance, Lewicki and Wiethoff (2000, p. 87) define trust as “an individual’s belief in, and willingness to act on the basis of the words, actions, and decisions of another,” while Rousseau et al. (1998, p. 395) describe it as “a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behavior of another.” It is also important to note that trust is often assumed to be inherently positive—as evident particularly clearly in recurrent public debates about the decline of trust, and in scholarly discussion of trust as an essential prerequisite of a functioning democratic society or civic community (e.g. Putnam, 1995). Yet, trust is valuable only in as far as it is placed in individuals, groups, or organizations that are trustworthy, and can be rather destructive if it is misplaced (Hosking, 2014; O’Neill, 2002).

The double-edged nature of trust, and its inherent link with vulnerability to the actions of others, is evident also in the case of trust in experts. Experts are, almost by definition, seen as trustworthy individuals. However, people can also place trust in individuals they might perceive as experts, but do not possess relevant knowledge and are therefore not trustworthy. Such situations have become increasingly common in the context of the contemporary information environment, where both the availability of information and the potential for exposure to misinformation has vastly increased (Hendriks et al., 2015). It is therefore important to consider how people establish the trustworthiness of experts, and specifically which indicators they use when doing so.

Studies that examine factors or indicators that inform public perceptions of experts use different conceptualizations, with some focusing on trust or trustworthiness (e.g., Besley et al., 2021; Hendriks et al., 2015) and others preferring to foreground concepts such as credibility (Fiske & Dupree, 2014). Regardless of these differences, existing research tends to agree that perceived expertise or competence is only one important indicator, and that people’s assessment also relies on factors associated with the expert’s behavior and character, such as perceived warmth, integrity, or good intentions (e.g. Besley et al., 2021; Fiske & Dupree, 2014; Hendriks et al., 2015). In developing the Muenster Epistemic Trustworthiness Inventory (METI), designed to measure laypeople’s evaluations of experts’ trustworthiness online, Hendriks and colleagues (2015) organized these factors into three groups: competence (expertise), adherence to scientific standards (integrity), and good intentions (benevolence), and an experimental study showed all three to be causally connected to trust in experts. In a recent study, Besley et al. (2021) suggested a slightly different operationalization and added a fourth dimension, namely openness, which refers to the scientist’s willingness to listen or to give others a voice.

In an increasingly mediated environment, how experts behave in media is also consequential for their perceived trustworthiness. Besley et al.’s (2021) addition of openness already points in this direction, as it captures an aspect of experts’ communicative behavior that has become much more visible in the context of online interactions. However, other aspects of communicative behavior, including choice of language, are also important; using aggressive language has been shown to decrease the trustworthiness of scientists participating in public debates (König & Jucks, 2019), while using technical language negatively affected perceived integrity and credibility (Thon & Jucks, 2017). The choice of communication channel matters, too; an experimental study showed that experts appearing on TV were perceived as more competent than scientists using YouTube, but at the same time, YouTube scientists were perceived as more entertaining and more easily understandable, which had a positive effect on trustworthiness (Reif et al., 2020). Furthermore, a panel survey experiment showed that scientists who post “selfies” from a laboratory environment are perceived as warmer and more trustworthy (Jarreau et al., 2019).

While factors discussed so far pertain to the qualities of individual experts—their competence, character, or their communicative behavior—it is also important to consider the specific communicative context in which experts operate. One important factor is the extent to which the specific subject of expertise has been a matter of uncertainty or controversy. For instance, research has shown that uncertainty has been used effectively to cast doubts on climate science (Oreskes & Conway, 2011), while ambiguity surrounding cancer prevention recommendations has been linked with lower perceptions of risk and lower adherence to preventive behavior such as regular screening or sunscreen use (Han et al., 2007). However, the effects of uncertainty differ depending on the cause of uncertainty—for instance, whether uncertainty arises due to technical issues or deficiency of methods used, or due to lack of consensus among experts. An experimental study conducted in the United States, examining perceptions of uncertainty across several issues from climate change to genetically modified food, showed that only the frame of consensus uncertainty is linked with significantly lower trust and belief in scientific claims (Gustafson & Rice, 2019). This is aligned with the conclusions of studies that have shown a link between increased public perceptions of scientific consensus on climate change and belief in climate change happening (e.g., Lewandowsky et al., 2012). It is also important to consider the contributing role of social media, which can magnify perceptions of controversy; indeed, attacks on scientists on social media have been shown to have a negative effect on perceived integrity and thereby trustworthiness (Gierth & Bromme, 2020).

Existing research surveyed so far identifies a range of pertinent indicators and factors that inform individual perceptions of expert trustworthiness. However, this body of work also has some limitations. First, it was conducted during periods of relative social stability and, hence, may not be able to account for all the factors that are at play in a crisis context. Existing research on past pandemics suggests that exposure to pandemics can lead to lower trust in science and scientists (Eichengreen et al., 2021), and some of the available research on trust in experts during the COVID-19 pandemic points in the same direction (Battiston et al., 2021). However, it is unclear whether these quantitative shifts in levels of trust also went hand in hand with qualitative changes in public perceptions of expert trustworthiness. It is therefore pertinent to ask the following research question:

• Research Question 1 (RQ1): Are the factors identified in existing literature—namely, the competence and character of experts, the presence or absence of expert consensus, and expert’s mode of communication—sufficient to explain how people establish expert trustworthiness in the context of a major public health crisis?

Second, existing research on factors informing public perceptions of expert trustworthiness is underpinned by the conceptualization of trust as a property of individuals, with limited consideration of the impact of systemic or societal determinants. However, trust can also be conceptualized as property of a social system, community, or society (Misztal, 1996), and comparative research on social trust (i.e. trust in other people) has shown that societies characterized by greater equality, higher levels of economic development, and greater quality of government tend to be more trusting (for an overview, see Ferrin & Gillespie, 2010). This raises the question of whether the factors informing perceptions of expert trustworthiness might also be affected by such systemic, cross-country differences.

Existing research on trust in experts is largely based on data collected in long-established liberal democracies of the West, which are relatively similar in terms of political systems and economic development and also have some of the highest rates of trust in science globally. As such, existing research has limited capacity to provide insight into the impact of systemic factors. That said, comparative research on the topic in the West does highlight the importance of systemic variables; for instance, a study of public trust in food biotechnology in 18 Western European countries and the United States (Priest et al., 2003) showed that the “trust gap” between trust in industry and trust in environmental and consumer organizations helps explain cross-country variations in acceptance of food biotechnology. At the same time, research examining determinants of individual-level trust in science and compliance with preventive behavior has shown that political and cultural variables play an important role. For instance, political ideology and religiosity have been shown to predict levels of trust in scientists (e.g., Brewer & Ley, 2013), low levels of trust in government have been linked to lower rates of acceptance of expert advice and adoption of protective public health measures (Blair et al., 2017; Dohle et al., 2020), and populism has been associated with distrust of science (e.g., Mede & Schäfer, 2020). It is therefore feasible to expect that independence from political elites will play a considerably more important role in countries that are affected by populism and where levels of trust in science and government are low. This raises the following question (RQ2):

• Research Question 2 (RQ2): Are the factors identified based on data collected in long-established Western liberal democracies sufficient to explain how people establish expert trustworthiness in countries that are marked by lower levels of trust in science, lower political trust, and populism?

Trust in Experts and Media Use

In existing research on trust in experts and science, media use is typically seen as one of the predictors of trust in experts, based on the assumption that media representations of experts and science shape public perceptions and thereby influence public trust. Studies conducted in the United States have established that heavy television viewing is associated with more negative attitudes to, and lower trust in, science (Dudo et al., 2011; Nisbet et al., 2002). In contrast, greater reliance on news media, newspapers, science television, and science magazines has been associated with higher levels of trust and more positive views (Anderson et al., 2012; Nisbet et al., 2002). In recent years, attention has shifted to the impact of digital and social media, with several commentators raising concerns about the potential detrimental impact of digital platforms and on the public status of expert knowledge (e.g. Nichols, 2017). Despite these concerns, research has shown a positive relationship between online media use, positive attitudes to science, and knowledge about science (Dudo et al., 2011; Huber et al., 2019). However, research conducted during the COVID-19 pandemic reveals a more troubling picture, with studies showing a link between lower levels of knowledge about the virus, or susceptibility to misinformation, and reliance on social media (Nielsen et al., 2020, p. 27; Xiao et al., 2021).

Arguably, the current, high-choice media environment brings new challenges for citizens seeking to find trustworthy information, partly because of the abundance of information, and partly because of the lack of editorial control. Through social media, citizens can get in touch with experts directly, and in such a context, trust itself can act as a determinant of media use, rather than being determined by media use. For instance, if people distrust expert information in mainstream media, they may be pushed to turn to digital media to find alternative voices. Indeed, existing research shows a link between lower levels of trust in mainstream news media and preference for nonmainstream news sources such as social media or blogs (e.g., Fletcher & Park, 2017). This requires us to conceptualize the relationship between trust in experts and media use as a two-way interaction, with media representations informing trust in experts, and trust in experts guiding the selection of media. This approach is also aligned with changes in science communication, which now combines the traditional, linear model of information flows, with a “networked” model within which scientists no longer have a monopoly on informing politicians and mass media, and a wider range of actors including nonexperts gains clout (Van Dijck & Alinejad, 2020). In our analysis, we therefore approach the relationship between media use and trust as a two-way interaction and ask:

• Research Question 3 (RQ3): How does trust in experts relate to people’s media use, and the way they access expert information?

Why Eastern Europe?

There are two key reasons that make a focus on trust in experts in Eastern Europe particularly attractive. First, Eastern Europe is among regions with the lowest rates of trust in science, as well as the lowest levels of vaccine confidence globally (Sturgis et al., 2021). This has been demonstrated again during the COVID-19 pandemic, with much of the region struggling with low support for vaccination, which was only around 50% in the Czech Republic, Hungary, and Poland (European National Panels, n.d.), and even lower in Serbia—arguably also a consequence of disinformation and conspiracy theories concerning COVID-19 and vaccines (Hajdu et al., 2020).

Second, despite the recent rise in trust in government in some Eastern European countries, including Poland and Hungary (European Commission, n.d.), levels of political trust in this region are lower than in established democracies (Závecz, 2017). This is important because trust in experts is—as noted earlier—known to be linked with levels of political trust (Blair et al., 2017; Dohle et al., 2020). Furthermore, governments in several parts of Eastern Europe were criticized for politicizing the pandemic and for abusing the state of emergency to extend their powers and restrict civic freedoms. Among the four countries covered in this study, this was particularly visible during the first wave in Hungary (Walker & Rankin, 2020) but also in Serbia and, to a lesser extent, in Poland (Guzek et al., 2021). In Hungary and Serbia, this led to frustration and distrust among some citizens and gave rise to starkly different responses to government handling of the pandemic among pro- and antigovernment citizens (Mihelj et al., 2022).

It is also important to note that the region experienced a marked decline in the quality of democracy in recent years—a development closely linked with the rise of populism (Vachudova, 2020). There has been significant democratic backsliding in Poland, and especially in Hungary, with governments criticized for eroding the rule of law and actively undermining democratic institutions, especially the judiciary and independent media (Vachudova, 2020; Surowiec & Štětka, 2020). The decline in the quality of democracy and the rise of populism have been observed in the Czech Republic as well, although the country still fares considerably better than Poland and Hungary (Hanley & Vachudova, 2018). In Serbia, democracy was never fully consolidated, and democratic standards have slipped again in recent years under the populist right-wing government and subsequent presidency of Aleksandar Vučić (Bieber, 2020). Given the growing acknowledgment of elective affinities between populism and distrust of science (e.g. Mede & Schäfer, 2020) as well as the recognition that populism can obstruct the successful management of public health crises (Hedges & Lasco, 2020), it is feasible to argue that the rise of populism in eastern Europe may be an additional factor contributing to distrust of experts in the region.

Methodology

Unlike most of the existing research on trust in experts, which is typically based on survey data, this study draws on a large qualitative dataset. This approach is consistent with the largely exploratory character of the study—namely, seeking to examine the way people establish expert trustworthiness in a crisis context, and in a sociopolitical environment that rarely features in existing research. The additional appeal of the dataset used here is associated with the fact that it was generated between February 2020 and May 2020, and therefore covers the first wave of the COVID-19 pandemic, including the introduction of stay-at-home measures in all four countries. Due to the nature of qualitative data collection, which typically requires a lengthy setup and planning period, it is rare to have access to qualitative data collected during the very start of a major crisis, as is the case here.

The dataset consists of semistructured qualitative interviews and diaries with participants from the four countries, 30 per country. Participants were recruited mostly from a quantitative survey conducted before the pandemic, ¹ which included a question about willingness to participate in a follow-up qualitative study. The participants were selected from the pool of volunteers identified via the survey, using the following criteria: (a) they had to be at least minimally engaged with news, meaning that they followed current affairs on a regular basis (minimum weekly); (b) using a quota approach, we sought to recruit equal numbers of men and women, three age groups (18–34, 35–59, and 60+), and locale (urban vs rural), and participants of different political persuasions (i.e. both pro- and antigovernment, determined based on voting patterns and opinions expressed in the survey). For hard-to-reach demographics that were not sufficiently well represented among survey volunteers (e.g., 60+), we recruited a small number of participants through personal contacts.

The first wave of interviews was conducted in February and March 2020, discussing topics such as media use, views on democracy, and opinions on social and political issues. The first interviews were then followed by media diaries kept for a 3-week period in March 2020, which coincided with the peak of the first pandemic wave and the introduction of lockdowns in all four countries. The diaries offer a detailed record of participants’ news habits during the early period of the pandemic, including their encounters with experts. The analysis of diaries informed the formulation of the second interview protocol, which provided an opportunity for discussing the impact of the pandemic on news consumption habits, the experiences of finding trustworthy information about the pandemic (including expert information), the perceptions of responses to the pandemic, and follow-up on selected diary items (see Online Appendix). These follow-up interviews were conducted remotely via video conferencing platforms in March and April 2020 and constitute the main basis of the finding presented in this article.

It is important to note that the interview protocol was designed to investigate news consumption and information-seeking behavior, rather being focused specifically on expert information and trust in experts. However, trust in experts emerged as a major issue of concern for participants, not only in response to direct questions about means of finding trustworthy information, including expert information, and about trust in experts (Online Appendix, questions 3a to 3d) but also in response to questions about misinformation and about trust in the government’s approach to the pandemic (Online Appendix, questions 5 and 9) and occasionally elsewhere in the interview as well. The interview protocol did not specify who an “expert” is, thereby allowing participants to adopt their own definition. This reflects the reality of people’s information-seeking behavior in the digital ecosystem, where people must rely on their own assumptions of what constitutes an expert.

Interviews, which lasted between 30 and 50 minutes, were transcribed from local languages to English, with transcripts ranging from 3000 to 8000 words. All material was analyzed using thematic analysis (Braun & Clarke, 2006) with support of the NVivo package. The coding tree largely followed the interview schedules and was formulated through a combination of deductive and inductive coding. Data collection and initial coding were conducted by four researchers fluent in local languages, with each of the four researchers coding the material from one country. As part of the initial coding, all four researchers identified all parts of materials in which experts were discussed, in both interviews and diaries. To ensure consistency and comparability, and resolve potential translation and coding problems, all interviewers used the same interview protocol (translated from English), attended a joint workshop prior to fieldwork, as well as regular online meetings during fieldwork and data coding. In addition, the project’s principal investigator read draft country reports (which included extensive excerpts of diary and interview material) ahead of each meeting and identified coding inconsistencies, which were then discussed and resolved during meetings, with country reports revised accordingly. In the second step of coding, a single researcher systematically examined all interview excerpts in which experts were mentioned, from all four countries, focusing on the identification of different reasons for trusting and distrusting experts, as well as the different information sources through which participants encountered experts. At several points in the coding process, the researcher consulted the principal investigator, who also read through a selection of interview material.

The study underwent relevant ethics checks and is compliant with Loughborough University’s Ethical Policy Framework and the Code of Practice for Research.

Results and Discussion

When examining participants’ answers to the question about trust in experts, two patterns stood out. First, participants from all countries, regardless of political orientation, stated that they trust experts. However, this general trust in experts did not extend to all experts. More specifically, in two of the countries—Serbia and Hungary—several participants expressed strong distrust of experts on government-appointed national crisis teams (i.e., Operative Body in Hungary and the Crisis Headquarters in Serbia). Second, the way in which participants justified their (general) trust in experts differed considerably from the way they justified their distrust of (some) experts. In the sections that follow we first examine justifications for trusting experts, followed by justifications for distrusting experts, and finally turn to the way trust in experts related to media use.

Establishing Expert Trustworthiness: Competence, Character, and Personal Contact

When talking about trust in experts in general, most participants had little to say, and simply stated that they trust experts because they are experts, without providing much detail. Those few participants who did elaborate further associated experts with greater specialist knowledge, experience, or integrity. Czech participant Cze-17 (male, 20) for example, explained that “these are people with some experience” while another Czech participant Cze-14 (female, 26) argued: “Well, they are experts, they should understand the matter, and so, you expect they will not issue false information.” Two Serbian participants explained that they trust experts because they know more about the virus (Srb-03, Srb-16), a Hungarian participant likewise noted that experts are knowledgeable about the topic (Hun-29), while a Czech participant explained his trust with reference to general trust in science and scientific facts:

I have always been oriented to science, I need the scientific data, I trust the people who work for many years in a profession, they are experts in their fields and their work can be checked. (Cze-04, male, 39)

Some participants explained their trust by contrasting experts with politicians, suggesting that experts were more trustworthy than politicians either because of their greater expertise, or greater integrity. One Polish participant, for example, emphasized that experts provided factual, disinterested information, while politicians provided only selected information with the aim of manipulating public opinion and emotions:

I believe the government wants to manage our emotional response by providing a specific type of information . . . they manipulate [information] in order to influence people and play on their emotions. Experts provide factual information and they leave room for my own interpretation and they allow me to draw my own conclusions. (Pol-19, female, 26)

The remaining factors discussed in the literature—namely, choice of communication channels and expert consensus—hardly featured at all in the discussion of (general) trust in experts. However, several participants specifically mentioned trusting experts they know personally, typically either their own doctor (Hun-02), or medical professionals among friends and family (Srb-07, Srb-22, Srb-24, Cze-04, Cze-09, Hun-04, and Hun-18). The following quotes from participants Hun-18 and Srb-24 illustrate this tendency well, and offer a sense of the range of people who qualified as “experts” in people’s eyes—from researchers with medical expertise to hospital employees:

I have two biologist/medical acquaintances and they wrote guides, and since they work with all kinds of bacteria as researchers, these acquaintances of mine, if they say about something that this is what people should do, then I say okay, I believe that. (Hun-18, male, 32)

I’m a layman, so I don’t know anything about the virus, just like those doctors don’t actually know a lot about the virus. But I have a friend who works in a hospital, and she sees people who have corona virus every day, and she knows all about their symptoms so she can tell me how to protect myself from it. (Srb-24, female, 26)

Reasons for Distrusting Experts: Lack of Independence From Political Elites, Lack of Expert Consensus, and Poor Choice of Communication Channels

In contrast to reasons for trust, reasons for distrust were more clearly articulated and involved more complex reasoning. The key reasons included all of the four factors discussed in the existing literature—namely, competence, character, consensus, and mode of communication. In addition, however, a large number of participants also mentioned the lack of expect independence from political elites. Another clear pattern emerging from the analysis was that the different factors did not operate in isolation; rather, participants’ answers show consistent patterns of interactions between them. First, and most prominently, the perceived lack of independence from political elites led several participants to perceive experts as less able to share their expert knowledge or as being in a position where their scientific integrity or goodwill was compromised. Second, for some participants, the perceived lack of expert consensus had a detrimental impact on perceived expertise. Third, the perceived poor choice of media outlet also led some participants to question experts’ competence, integrity, as well as independence.

The perceived influence of political elites on experts was a particularly prominent reason for distrust in Hungary and (to a somewhat lesser extent) in Serbia. In both countries, several participants explained that they distrust experts on the government-appointed national crisis teams, primarily because they were perceived as lacking in independence, which then shaped the perception of their ability to exercise professional judgment. Members of the Operative Body were described as soldiers not medical experts (Hun-27), as not credible (Hun-30), or not actually experts (Hun-05, Hun-12, and Hun-18), lacking in independence (Hun-05), only saying what they are allowed to say (Hun-10 and Hun-22), fulfilling a political order (Hun-06), not telling the truth (Hun-07 and Hun-12), and changing their opinion depending on context (Hun-11). The following excerpt elaborated these reservations in relation to a prominent member of the Operative Body, Cecília Müller, a physician who also serves as the country’s Chief Medical Officer.

This is obvious, an expert is an expert, a politician smooths things over and propagates, but an expert . . . if they are not influenced by others . . . if their opinion is not influenced by others, then they are independent and have a professional opinion. You know? But who are the experts? The problem here is that the experts are politically inclined, and they form their expert opinion to align with politics. Then we should ask where can we find an independent expert whose opinion is acceptable? And whose opinion is right and that I should follow? Tell me, which expert is the real expert? . . . Cecília Müller is an expert, because she is a doctor, but she is not independent politically. Then? . . . This is a good question, too, everyone is an expert, but if they motivate me more, then I’ll be most expert of all. So . . . anyway, again, this is not politics . . . But I would trust experts, because they are supposed to be professionals. (Hun-26, male, 64)

For one Hungarian participant (Hun-07), reservations about the Operative Body and its close ties with the Government were so strong that he felt compelled to trust an expert simply because their opinions were openly rejected by the Operative Body:

As I said, Cecília Müller, the Chief Medical Officer, said that only those who are sick have to wear a mask. And now, um . . . János Szlávik, the chief doctor of the Szent László hospital said that everyone should wear a mask, and he is the most important doctor working in epidemiology in Hungary, he said that we should wear it . . . And I believe Professor Szlávik more than Cecília Müller, because Professor Szlávik was there at the first meeting of the Operative Body, at the first interview, and they never invited him again, he disappeared . . . I wonder why. (Hun-07, male, 52)

Several Serbian participants expressed similar distrust for members of their Crisis Headquarters. As participant Srb-26 explained: “I realized that certain doctors did not have their own opinions, and that they were waiting for those in power to give them a green light when someone would ask them a question.” Other reasons mentioned by Serbian participants included: the experts are not really experts, or at least were not chosen for their expertise (Srb-06); the doctors are told what to say (Srb-07); they always spin the same stories (Srb-21); and they seem to be keeping something from us, not telling us the whole truth (Srb-28).

In Poland and the Czech Republic, concerns about political influence over experts were marginal, although one Czech participant (Cze-14) did argue that politicians fired qualified experts from the country’s Crisis Staff due to political differences. Although we should be aware of the limits of extrapolating cross-country differences based on a small sample, it is worth noting that these cross-country differences are consistent with existing findings on citizen responses to the handling of the pandemic and can be linked to higher levels of politicization of the pandemic in Serbia and Hungary at the time (Mihelj et al., 2022).

The second most common reason for distrusting experts was disagreement among experts (Consensus). Lack of expert consensus led some participants to be confused over who to believe, and thus distrustful—a finding aligned with existing research on the impact of scientific consensus on trust (Gustafson & Rice, 2019; Oreskes & Conway, 2011). This reason was particularly often mentioned in the Czech Republic. For example, Czech participant Cze-15 (female, 44): “Physicians themselves do not agree on their views. Some of them say flu is worse, I don’t know. It’s hard to say.” Czech participant Cze-11 explained that he followed the opinions of medical experts from the beginning of the pandemic, but quickly lost faith due to their contradictory opinions, and specifically due to the perceived conflict between proponents of lockdown measures and proponents of the so-called herd immunity strategy:

I was surprised that there are many experts with different opinions and, frankly speaking, I am tired of it. I think that people without education must be very puzzled of it . . . And now we have here this pandemic and half of epidemiologists say let’s keep lockdown and the other is for letting the virus go into population. So, what is a common man to think? Maybe, it is better to pray. . . [I followed them] from the beginning but then I found out there are many contradictory opinions. Thousands of doctors and experts and it is like lawyers . . . The crisis showed that we cannot believe them, such a disagreement in opinions, experiments . . . (Cze-11, male, 34)

For some participants, distrust arose due to the way in which experts performed during media appearances, or due to the choice of media in which they appeared. For instance, Serbian participant Srb-21 (female, 21) mentioned distrusting national experts because “they don’t know how to answer the journalists’ questions all that well.” Another participant from Serbia mentioned distrust in experts who appeared in an untrusted news source (i.e., commercial TV channel Happy):

I was watching both the doctors’ conferences at 3PM, and the government’s conferences at 8PM. But then I stopped when after one conference all the doctors went to Happy and spent more than three hours on one of their talk shows. I realized it was all a parade and that I shouldn’t bother with it anymore. (Srb-06, female, 31)

In Poland, conservative and pro-government participant Pol-05 distrusted experts who provided evidence to antigovernment commercial TV stations like TVN, arguing that such experts are more politicians than real experts:

I trust the politicians from the government more than the experts, but that also depends on what type of experts you’re talking about. The doctors, I trusted what they said. When it comes to experts shown on TVN or other channels, or some show-off doctors who were complaining that things are handled better in other countries, I was very mistrusting. If things were really handled better elsewhere, then why did things turn so bad in the end? I felt that what these self-proclaimed experts say is grossly exaggerated.” (Pol-05, male, 51)

This last quote also shows how trust in experts can be influenced by the nature of the political and media context, and by people’s own political position in this context. The Polish media landscape is heavily polarized, with the progovernment public service broadcaster TVP and the antigovernment commercial broadcaster TVN typically providing contrasting versions of current affairs. The audience is largely split along similar lines, and as our data showed, some participants judged the independence and hence trustworthiness of experts based on whether they appeared in media aligned with their political convictions.

Trust in Experts and Media Use

Most participants, regardless of their views on experts and across all countries, tended to encounter experts through government communication on COVID-19, especially in the form of live government briefings that featured experts. These live events were followed either on television or radio or through live streams via digital platforms. News reports featuring experts were important as well, accessed through television, radio, digital news platforms, and (less often) print media. Several participants from all countries recalled following daily government briefings, during which representatives from the ministry of health or national crisis teams provided updates on the number of new cases and the latest preventive measures. For instance, participant Pol-27 commended the conduct of the Polish Minister of Health Szumowski during these conferences:

For example, The Ministry of Health. I’m not particularly fond of Prime Minister’s press conferences nor am I a fan of his oratory skills. Szumowski—the Minister of Health—on the other hand, definitely seems trustworthy. The way he conducts himself during press conferences, the way he speaks, can seem very reliable. (Pol-27, male, 66)

Also common were mentions of official COVID-19 webpages, maintained by relevant ministries or health authorities, and interviews with experts encountered as part of regular news bulletins on radio or television, as for instance with Czech participant Cze-13, who recalls TV programs featuring the Czech Deputy Minister of Health, Prymula:

For example, in Česká televize, I watch the main news, Události. I do not think the news should vary in other media. And after the news, there are various programs with experts, for example with Prymula and the like. (Cze-13, female, 39)

In contrast to mainstream media, which were used as a key source of expert opinion in an equal measure in all countries, there was notable variation found in the use of social media, specifically YouTube and Facebook. Namely, the use of social media as sources of expert opinion was particularly common in Hungary and in Serbia. Given that Hungary and Serbia were also the two countries where several participants distrusted the local crisis teams, this pattern suggests that distrust in government-appointed experts (who typically featured in mainstream, progovernment news media) may have led citizens to rely on social media for expert opinion.

Our participants’ comments confirmed this was the case, although the types of experts found on social media varied. On the one hand, several participants turned to social media to look for foreign experts; these participants tended to be antigovernment and university-educated and looked for experts they perceived as trustworthy because of their expertise, assessed through their professional positions and institutional affiliations. For instance, participant Hun-21 (male, 31) reported generally checking YouTube for information about COVID-19, seeking out experts in research positions from trusted foreign organizations: “Well, whatever popped up due to YouTube’s algorithm, once it was Chicago University’s leading doctor talking, I will believe him. I look at WHO or whatever YouTube brings up.” In a similar vein, Serbian participant Srb-21 (female, 21) also distrusted national experts, and this led her to compare the answers of Serbian experts to those of foreign experts, which she accessed through YouTube. Participant Hun-22 was likewise distrustful of Hungarian experts, and sought foreign information on COVID-19 through Facebook, usually in the form of English language articles from trusted U.K. news sources.

Yes, social media . . . Not specifically by searching for it, but if I see that there is this kind of article about, I don’t know, the UK or someone living in the UK, because I think the media is much freer there and that the doctor is able to say what they think and what they truly experience, because I think . . . they are intimidated. (Hun-22, female, 36)

This excerpt also reveals that the participant used the perceived independence of the media as a means of assessing expert trustworthiness. U.K. media are implicitly contrasted with Hungarian media, which are perceived as lacking in independence, and this increases the perceived trustworthiness of experts in U.K. media.

Not all participants who distrusted national crisis teams turned to foreign experts. Several searched for domestic experts who did not appear in mainstream news, either because they were not among the experts chosen by the government (Hun-18, Cze-25), or because they were banned by the government (Pol-25 and Srb-06). For example, Polish participant Pol-25 explained that she was attracted to a Polish expert because he was purportedly banned from sharing his opinions as he was critical of government’s actions. This participant also noted the expert’s professional status as a consultant as an indication of expert status and his appearance in media that allowed him to air his critical views:

For example, yesterday there was one with Professor Simon from Wrocław, who, I believe, is a national consultant. Other national consultants, I guess 89 of them, were also banned from pronouncing their opinion on the coronavirus. And Professor Simon said, “All right, I am banned from stating my opinion, so now I am going to present my view as a regular doctor.” Now he is invited to participate in various TV programs and airs his views on how things look. (Pol-25, female, 79)

Serbian participant Srb-06 likewise stated that they trusted those experts who were not prominent in the media, indicating that lack of prominence in (government-controlled) media was a marker of expert independence and hence trustworthiness:

I also watched certain TV shows which I found on YouTube, and which dealt with what certain virologists and epidemiologists, who weren’t prominent in the media, had to say about the Coronavirus. I trusted them more than the doctors from the Crisis Headquarters. (Srb-06, female, 31)

However, some participants who were driven to social media due to distrusting government-appointed experts were less able to identify reliable sources of expert opinion. Some fell prey to untrustworthy experts—typically people with some background in natural sciences but not specifically virology or epidemiology, who shared misinformation. For example, two Hungarian participants mentioned Doktor Gödény, a Hungarian pharmacist and fitness personality who was adamantly against mask-wearing and lockdown measures. He often posted his opinions questioning preventive measures against COVID-19 and was later charged by Hungarian authorities for spreading disinformation.

What also emerges from our analysis is the widespread reliance on YouTube as a source of expert opinion. Our participants’ answers indicate that the attraction of the video-sharing platform lay in the combination of audio-visual communication and the absence of editorial control, which resulted in a perception of greater authenticity and directness. Some participants appeared to turn to YouTube and other social media to look specifically for expert information that reinforced pre-existing opinions. For example, Hungarian participant Hun-15 (male, 38) believed that the measures imposed by the government were excessive, and followed the already mentioned Doktor Gődény on YouTube: “Yes, because I think it would be enough to isolate the vulnerable age groups and those with chronic illness and the others should be allowed to continue producing the GDP. It will be much worse if the economy crashes than if the virus spreads.”

These examples offer a good illustration of the double-edged nature of social media as sources of expert information, which becomes particularly pronounced in politically polarized contexts, or in relation to contested topics. In a context where experts’ trustworthiness is compromised due to perceived political influence, social media can enable access to trustworthy expert information, but can also offer access to misinformation, and further polarize and politicize the discussion, thereby contributing to distrust.

Conclusions, Limitations, and Recommendations for Future Research: Toward Examining Trust in Experts as Part of a Nexus of Trust

The results of our study bring several contributions to existing knowledge on trust in experts. First, in response to RQ1, the results highlight two additional factors that inform public perceptions of expert trustworthiness, which are largely absent from existing literature on the topic: personal contact with (perceived) experts and expert’s independence from political elites. It is feasible to argue that both factors are more likely to be important in a context of crisis. A crisis such as the COVID-19 pandemic increases a sense of uncertainty and raises doubts over who to trust (Jennings et al., 2021), and in such a context, people are arguably more likely to turn to people they know personally. That said, personal contact with experts can potentially increase trust even outside of a conflict-ridden context. For instance, a comparative study of trust in energy experts among homeowners in Belgium and Denmark has shown that the Belgian system, which provides the homeowner with a personal contact with experts, is conducive to higher levels of trust in experts (Gram-Hanssen et al., 2007). Taken together, the results of this study and our own research suggest that personal contact should be more systematically incorporated into theoretical models and empirical studies of trust in experts.

Independence from political elites can also assume greater importance in a crisis-ridden context, especially if it is accompanied by a distrust of political elites, which was the case in some of the countries we studied. This also highlights the importance of considering the impact of sociopolitical context on trust in experts and brings us to RQ2. The material presented here confirms that factors identified based on data collected in long-established Western liberal democracies may not be sufficient to explain how people establish expert trustworthiness in different sociopolitical contexts. Especially in countries affected by populism, and where levels of trust in government and trust in science are low, independence from political elites is likely to play a key role in shaping people’s assessment of expert trustworthiness.

Let us add that the political independence of experts may also play a role in countries with higher general levels of political trust and trust in science—although the link may only become apparent when considering different types of scientists or when looking at scientific knowledge that is a matter of public controversy. For instance, polls conducted in the U.K. show that levels of public trust in scientists vary depending on their employer, with scientists working for universities perceived as considerably more trustworthy than those working for governments (Krause et al., 2019). A recent evaluation of existing research on factors affecting public reception of climate science also highlighted the importance of independence as a contributing factor (Hahn et al., 2016). In sum, these studies, along with our own results, suggest that the political independence of experts should be included in future studies of trust in experts, and examined both at a systemic, cross-country level and at an individual level.

Another important finding emerging from our analysis, relevant to both RQ1 and RQ2, relates to the interactions between factors that shape public judgments of expert trustworthiness. Most prominent was the interaction between the perceived lack of independence from political elites and expert’s character and competence, with several participants doubting experts’ competence and/or integrity because they felt they lacked independence from politicians. In some cases, this interaction extended to people’s assessment of expert’s handing of public communication; if an expert appeared in a media channel that the participants perceived as politically biased, or out of line with their own convictions, this negatively affected perceived trustworthiness. This last point also brings into play yet another form of trust—namely, trust in the media. To put it differently, our results suggest that trust in experts interacts not only with trust in political elites but also with trust in the media, thereby forming a complex nexus of trust.

Third, in response to RQ3, our analysis offers new insights into how trust shapes people’s information seeking online. When examining media use in relation to experts, our analysis showed a higher level of reliance on social media in countries with higher distrust in the local crisis teams. When participants felt that they could not access trusted information from mainstream channels, which are often the only ones widely available in these countries, they turned to online sources, often to seek out foreign experts, with some also turning to untrustworthy individuals spreading misinformation.

These findings have implications for scientists, science communication practitioners, and professionals participating in health crisis communication, as well as for media regulators and journalists involved in reporting on science and/or covering a health crisis. The crucial role of political independence suggests that experts should seek to maintain professional independence and resist attempts to politicize expert knowledge. Related to this is the importance of careful consideration of suitable channels of communication. Especially where expert knowledge becomes an object of political polarization (as was the case with expert knowledge informing lockdown restrictions and mask-wearing mandates during the COVID-19 pandemic) priority should be given to media outlets with an established reputation for impartial and professional reporting. The importance of personal contact also suggests that developing more localized, community-based communication strategies, and seeking the involvement of local experts in community-based public awareness campaigns, may help increase the effectiveness of messages. Finally, trustworthy sources of authority with an established online presence could be recruited to contribute to online campaigns, including campaigns specifically designed to counter misinformation.

Regulatory authorities in the sphere of media and communication should seek to develop support and guidance for media organizations during public health crises and similar events, and such guidance should also include recommendations on how best to support preventive measures in a context where political elites might be abusing the crisis for political gain. Regulatory authorities should also strive to act as a protector of freedom of media independence and access to information during a crisis and challenge the attempts of political authorities to interfere in the circulation of information. Finally, journalistic associations, as well as individual media organizations (both public and commercial), should develop guidance for journalists’ conduct during a public health crisis. Special attention should be paid to guidance for reporting on divisive public health measures, and guidance on communicating scientific findings and expert advice, especially in a context where political elites may be seeking to interfere in expert work, and/or where expert knowledge becomes politicized. ²

We should also acknowledge some limitations of our research. First, our study covers a limited period early in the pandemic, and the dynamics of trust formation may have shifted since, following changes in the political context. Second, we focus on general trust in experts, rather than differentiating between general trust in science and/or scientists and trust in different types of sciences and expert knowledge. However, trust in experts can vary depending on the type of expertise involved; for instance, while general trust in science may be high, trust in specific types of sciences may vary, especially in case of controversial topics such as climate change or nuclear energy (e.g., Krause et al., 2019) Third, our results suggest that public trust in experts is only one piece of a complex nexus of trust, which shapes the capacity of our societies to draw on expert knowledge and use it effectively to respond to a health crisis or other societal challenges. Through our research, we were able to highlight some of the important components of this nexus—namely, trust in the government and trust in the media. However, it is possible to envisage this nexus of trust extending to, for instance, public trust in religious authorities, courts, industry, or even specific political pressure groups or civil society actors (cf. Priest et al., 2003) as well as to trust in science among policymakers and government trust in the public (Cairney & Wellstaed, 2021). Future research should pay closer attention to this wider nexus that trust in experts is embedded in and examine how it varies from country to country.

Supplemental Material

sj-docx-1-scx-10.1177_10755470221100558 – Supplemental material for Establishing Trust in Experts During a Crisis: Expert Trustworthiness and Media Use During the COVID-19 Pandemic

Supplemental material, sj-docx-1-scx-10.1177_10755470221100558 for Establishing Trust in Experts During a Crisis: Expert Trustworthiness and Media Use During the COVID-19 Pandemic by Sabina Mihelj, Katherine Kondor and Václav Štětka in Science Communication