Skip to main content
NCBI home page
Sage Choice logoLink to Sage Choice
. 2025 Jun 5;34(8):1046–1065. doi: 10.1177/09636625251337709

Mediating trust in content about science: Assessing trust cues in digital media environments

Justin T Schröder 1,✉, Lars Guenther 2
PMCID: PMC12535621  PMID: 40470595

Abstract

Intermediaries such as (digital) media use trust cues in their content, that is, information and linguistic markers that present public audiences reasons for trusting scientists, scientific organizations, and the science system. Trust cues refer to dimensions of trust in science such as expertise, integrity, benevolence, transparency, and dialogue. Because digital media environments are expected to be heterogeneous in content, the sources of trust cues, characteristics of objects of trust in science (e.g. the gender of scientists), and their impact on public trust in science may vary. In our quantitative content analysis, we identified trust cues across several sources of scientific information (n = 906) and examined their heterogeneity in digital media environments. Our results reveal journalism as the most important source for trust cues and that scientists are the most prevalent object of trust—with female scientists being underrepresented. Differences across (digital) media imply varying impacts on public trust in science.

Keywords: digital media environments, intermediaries of trust, quantitative content analysis, trust cues, trust in science

Scientific information, applied when using technology, making medical decisions, and considering ecological trends (e.g. Hendriks et al., 2015), is important in people’s everyday decision-making. Because publics are somewhat dependent on science and scientific information, their trust in science is among the most important variables to engage with such information (e.g. Plohl and Musil, 2021; Saffran et al., 2020). Today, most people get in contact with science and receive scientific information through (digital) media (e.g. European Commission, 2021; Wissenschaft im Dialog, 2023), including in Germany, where nearly the entire population uses the Internet (e.g. ARD/ZDF–Forschungskommission, 2023), and online media is an important source of scientific information (e.g. Wissenschaft im Dialog, 2023). Given those trends, (digital) media are intermediaries of trust in science, and in that role, they mediate trust via trust cues. When it comes to science, trust cues are information, language, and linguistic markers used in content that provide public audiences, as subjects of trust, reasons to trust in science (e.g. Schröder et al., 2025). They enable public audiences to evaluate whether they should trust scientists, scientific organizations, and the science system as objects of trust. Because media content can indeed evoke trust in science (Guenther et al., 2024), in this article we use the term trust cues as shorthand for trust-evoking cues. 1

At a time when scholars are discussing and fearing a decline in public trust in science (e.g. Kennedy and Tyson, 2023) and when findings of media’s effects on trust in science remain mixed (see also Guenther et al., 2024), 2 it seems beneficial for researchers to focus on specific aspects of media content—for instance, the trust cues used. Although scholars have yet to focus on such cues, studies on audiences have provided hints of their effectiveness. For example, research has shown that communicating open science practices, which are directly linked to transparency, positively affects trust in science (e.g. Rosman et al., 2022; Song et al., 2022) and that assessments of trust improve when scientists correct themselves and communicate uncertainty (e.g. Altenmüller et al., 2021; Ratcliff and Wicke, 2023). Because those insights suggest that trust cues can indeed evoke trust in science, the intermediaries of such trust deserve a closer look.

Intermediaries of trust especially warrant attention in digital media environments, where a heterogeneity of actors and content means that diverse interests can access diverse audiences (e.g. Brossard and Scheufele, 2022; Huber et al., 2019; Schäfer, 2016). In digital media environments, not only professional journalists but also non-journalistic actors communicate about scientific issues and therefore become relevant sources of content about science (e.g. Taddicken and Krämer, 2021; Weingart, 2017). For the same reason, audiences face the risk of receiving misleading or false information, particularly in populist, non-mainstream media (e.g. Frischlich et al., 2022) and social media (e.g. Jennings et al., 2021; Xiao et al., 2021) compared with journalistic media. For that reason, digital media environments have been serving as an explanation for declining public trust in science (e.g., Weingart and Guenther, 2016). That means that when analyzing trust cues in content about science, differentiating professional journalistic, populist/non-mainstream, social, and other Internet-based media (e.g. blogs) is crucial to capturing the heterogeneity of actors and content in digital media environments.

The same heterogeneity also affects objects of trust, especially scientists and their representation in terms of gender. To date, research has generated mixed findings concerning scientists’ gender and trust in science. For instance, some studies have demonstrated that trust assessments indicating expertise are associated with male scientists due to stereotypes, whereas others have revealed that participants tend to have higher trust in female than in male scientists (e.g. Hubner and Bullock, 2024). Owing to those discrepancies, the gender of scientists should be included in analyses of trust cues, for it can afford additional insights into how objects of trust are represented.

Altogether, investigating how trust cues in scientific content vary across different (digital) media (i.e. journalistic, populist, social, and other Internet-based media) and based on scientists’ gender, can clarify how those factors may affect public trust in science. A deeper, more detailed analysis of media content could also illuminate recent developments regarding public trust in science. In those ways, such research can form a foundation for further analyses of media’s effects on trust in science and elucidate the trust relationship between science and digitized publics. Thus, in our study, we asked the following overarching research question: How does the use of trust cues in content about science vary across different (digital) media and genders of the scientists represented?

1. Theoretical background

Trust is a multifaceted concept that has been examined in various fields of study and thus has numerous definitions. In this article, we view trust from a sociological-theoretical perspective within communications research (e.g. Reif and Guenther, 2022; Schröder et al., 2025), which characterizes trust as a mechanism for reducing complexity (e.g. Kohring, 2016; Luhmann, 2014). For instance, when a subject of trust has trust in an object of trust, it means that the subject has confidence in the object to achieve future tasks and/or expectations and, in turn, is willing to risk the possibility that such tasks and/or expectations might not be achieved (e.g. Giddens, 1990; Kohring, 2016). In that way, the subject makes itself vulnerable (e.g. Goldenberg, 2023; Mayer et al., 1995; Reif and Guenther, 2022; Resnick et al., 2015). However, trust is crucial for dealing with complexities in modern societies (e.g. Luhmann, 2014). In the context of our research, focused on public trust in science, an especially valuable concept is epistemic trust, which means depending on science as a producer of reliable knowledge, both its inherent validity and its perception as a secure, dependable source of information (e.g. Origgi, 2012; Sperber et al., 2010).

Trust cues in content about science

Bentele’s (1994) theory of public trust, which recognizes media’s role in the trust relationship between science and its publics, 3 suggests that public trust has emerged as a distinct form of trust within the public sector. According to the theory, trust in individuals, organizations, and systems is influenced by information conveyed through media, including facts, events, and messages (e.g. Bentele, 1994; Reif, 2021). Our conceptualization of trust cues aligns with Bentele’s theory of public trust (1994) insofar as it focuses on media content (for an overview, see Schröder et al., 2025) and emphasizes the importance of content about science. By extension, trust cues are conceived as influential linguistic components in content about science that present public audiences with reasons to trust in science, specifically by addressing established dimensions of trust in science. Thus, trust cues are an operationalization of those dimensions. 4

When it comes to trust in science, multiple dimensions have been distinguished, including the well-established dimensions of expertise, integrity, and benevolence (e.g. Hendriks et al., 2015, 2016; Mayer et al., 1995). Albeit initially developed for science at the micro (e.g. Hendriks et al., 2015) and meso levels (Mayer et al., 1995), their applicability at the macro level has also been conceptualized (e.g. Reif and Guenther, 2022) as well as empirically tested and validated (Reif et al., 2024). Moreover, in recent research, new dimensions have been suggested, including openness (e.g. Besley et al., 2020) and, considering the reciprocal concept of dialogue behind public engagement with science, transparency and dialogue (e.g. Reif and Guenther, 2022). Indeed, Schröder et al. (2025) included transparency and dialogue, which can also be applied at all three levels, in developing the concept of the trust cues. Because all dimensions of trust in science can be referred to by trust cues, we refer to them as expertise, integrity, benevolence, transparency, and dialogue cues, respectively. For each dimension, respective trust cues were identified by Schröder et al. (2025) in a qualitative pilot study with a diverse media sample. The result of this study was a list of 35 trust cues identified, with definitions and examples.

In this regard, expertise illustrates science’s capacity to identify, evaluate, and target problems by using specialized knowledge acquired through education, experience, and qualifications (in the respective fields of research). Thus, expertise cues refer to mentions of the academic education, professional experience, and qualifications possessed by a scientific object of trust, including their academic degree(s), organizational affiliation, and/or their reputation (e.g. Hijmans et al., 2003; for all definitions see also Reif et al., 2024). Integrity means the assurance of objectivity, validity, and reliability achieved by adhering to scientific standards and processes. As such, integrity cues refer to mentions of independence (e.g. funding sources), assurances of scientific quality (e.g. uncertainties in research and peer review), and references to scientific standards and processes in general (see also Guenther et al., 2019). The dimension of benevolence refers to science’s orientation toward ethical norms and moral values as well as awareness of its responsibility to society (see also Hendriks et al., 2015). In that regard, benevolence cues address ethical norms, social responsibilities (e.g. in researchers’ communication of research-related risks or their assessment of public events and current affairs), and benefits for society that science might provide (e.g. science-based recommendations). Transparency refers to the expectation that scientific actors should make research and related scientific information publicly accessible; by extension, transparency cues refer, for example, to comprehensible language that also includes technical terminology and the accessible presentation of research results (e.g. Reich, 2011; Reif, 2021). Concerning the dimension of transparency, it is important for science to be transparent regarding its research and not necessarily for a piece of science-related information to be communicated transparently. For example, if journalists mention a research project’s funding source, this information is communicated transparently (by journalists, not by scientific actors) but does not allow public audiences to evaluate science itself being transparent but rather to evaluate its independence. Dialogue refers to science that participates in and enables interaction with public audiences (e.g. Reif et al., 2024). Thus, dialogue cues mean the participation of scientific objects of trust in public events, their journalistic media presence and direct media presence (e.g. use of social media), and their public engagement in research. Dialogue presupposes a certain form of reciprocity (i.e. questions can be asked, and interaction can occur). Such reciprocity distinguishes dialogue from transparency, which refers to science being transparent but without considering interactions.

Taken together, all dimensions can be addressed by actors at the macro, meso, and micro levels (e.g. Reif and Guenther, 2022; Schröder et al., 2025). In our study, we investigated the extent to which trust cues can be identified at each level. Apart from our study, few content analyses have (indirectly) considered those five dimensions of trust in science or analyzed linguistic markers that can be interpreted as trust cues (e.g. Welzenbach-Vogel et al., 2021; for an overview see Schröder et al., 2025). Table 2 presents an overview of all trust cues that we examined, their respective superordinate categories, and the dimensions of trust in science that they refer to (see also Schröder et al., 2025). These trust cues have not been quantified yet to examine their prevalence across (digital) media.

Table 2.

Overview of trust cues, superordinate categories, and dimensions of trust in science.

Dimensions Categories Trust Cues Sample
n %
Expertise 2660 44.8
Academic education 53 0.9
Professional experience 28 0.5
Qualification* 2579 43.5
Academic degree 185 3.1
Reputation 78 1.3
Professional position 257 4.3
Affiliation to an organization 1185 20.0
Department or area/discipline of expertise 874 14.7
Integrity 1363 22.5
Independence 82 1.4
Client 27 0.5
Funding source 37 0.6
Interests 18 0.3
Scientific quality assurance 318 5.4
Correction/Revision 19 0.3
(Un)Certainties (& Limitations) 226 3.8
Peer review 14 0.2
Continuity/Permanence of research 58 1.0
Scientific standards and processes 963 16.2
Legal framework for research 11 0.2
Description (and explanation) of research processes 513 8.6
Working conditions in science 16 0.3
Research collaboration 73 1.2
Publication 351 5.9
Benevolence 1332 22.5
Ethical norms 19 0.3
Social responsibility 679 11.4
Research-related risks 5 0.1
Prediction 176 3.0
Assessment of public events/current affairs 498 8.4
Benefits for society 634 10.7
Social significance of science 19 0.3
Discoveries and breakthroughs 144 2.4
Applicability of results 240 4.0
(Science-based) recommendations 197 3.3
Personal reasoning for benevolent behavior 34 0.6
Transparency 373 6.3
Accessibility of results 46 0.8
Comprehensible language 327 5.5
Dialogue 204 3.4
Participation in public events 17 0.3
Media presence 175 3.0
Journalistic media presence 104 1.8
Direct media presence 23 0.4
Further media presence 48 0.8
Public engagement in research 12 0.2
Open in a new tab
*

Frequencies of trust cues and categories of trust cues with a distribution ⩾5% are in bold.

Heterogeneity in (digital) scientific content

With the Internet and digitalization, a wide range of communicators have gained access to (digital) public spheres that allow them to bypass journalistic selection criteria (e.g. Pavlik, 2000). Consequently, a variety of actors can now express themselves online and thereby reach public audiences (e.g. Brossard, 2013; Schäfer, 2016; Schröder and Guenther, 2024; Taddicken and Krämer, 2021). For that reason, digital media can facilitate greater heterogeneity in content, the sources of trust cues, and objects of trust in science.

In this article, heterogeneity in terms of content is defined from the perspective of trust as a multidimensional concept, such that scientific content can refer to expertise, integrity, benevolence, transparency, and dialogue expressed via a range of trust cues. Aside from journalists, various science communicators have emerged online to address scientific topics using multiple sources with diverse interests. In social and other Internet-based media, science public relations professionals, political and governmental actors, bloggers, and other communicators with broad influence online—for instance, influencers—may discuss such topics. In populist media, by contrast, alternative experts and state propaganda may emerge (e.g. Weingart, 2017), and online sources may thus share misleading and/or false information, as seems to be more prevalent in populist, non-mainstream media (e.g. Frischlich et al., 2022; Taddicken and Krämer, 2021) and on social media (e.g. Jennings et al., 2021; Xiao et al., 2021) than in journalistic media. Last, heterogeneity in objects of trust in science encompasses science performed at multiple levels, including scientists (i.e. micro level), scientific organizations (i.e. meso level), and the science system (i.e. macro level). Recognizing the heterogeneity of (digital) media environments and the complexity of trust, in our first research question (RQ1) we asked: How does the use of trust cues differ across (digital) media with respect to (a) the dimensions of trust in science, (b) science as the object of trust at different levels (i.e. micro, meso, and macro), and (c) their sources?

Scientists play an especially important role in journalistic media because journalists, in their coverage of science, typically search a human angle, which is also a news factor in science journalism (e.g. Amend and Secko, 2012; Guenther, 2019). In light of that trend, scientific actors at the micro level may also be important in other types of media. Concerning how scientists are represented, previous research has revealed that female and male scientists are not equally represented in science media coverage (e.g. Fletcher et al., 2021; Kitzinger et al., 2008; Mitchell and McKinnon, 2019; Niemi and Pitkänen, 2017), in terms of not only frequency (e.g. citations and voices heard) but also certain characteristics (e.g. stereotypes; Chimba and Kitzinger, 2009; GMMP, 2020; Joubert et al., 2022; Kitzinger et al., 2008; Mitchell and McKinnon, 2019). That imbalance is negative in critical ways. After all, although women contribute significantly to scientific progress and gender diversity is supposed to drive innovation (e.g. Hofstra et al., 2020), 5 female scientists are significantly underrepresented compared with their male counterparts in the coverage of science (e.g. GMMP, 2020) in various types of media (Kitzinger et al., 2008). In the context of trust in science, initial research has revealed qualitative similarities and differences between female and male scientists. For instance, when describing a scientist’s expertise, their qualifications were in the spotlight regardless of their gender. However, the personal biographies detailing scientists’ motivations in their work were provided only for female scientists, whereas scientific advice conveying their benevolence toward society was provided only for male scientists (Schröder, 2025). On that basis, quantitative gender differences in the context of public trust in science can be assumed.

Differences in gender representations are also tied to journalists’ selection of sources and framing processes (e.g. Niemi and Pitkänen, 2017) as well as to their own gender. Gender aside, journalists covering science predominantly cite male scientists while limiting their references to female scientists (e.g. GMMP, 2020; Kitzinger et al., 2008); however, female journalists choose sources who are women significantly more often than male journalists do (GMMP, 2020). Thus, taking journalists’ gender into account might provide additional insights into how it impacts the representation of gender in science coverage.

Considering all of the above, in our study we hypothesized that gender accounts for the heterogeneity of scientific objects of trust in science as well as in selecting sources of science coverage, which seems important for analyzing trust in science. For additional insights into that issue, in RQ2 we asked: How does the gender of the object of trust in science at the micro level relate to (a) the dimensions of trust in science, (b) the trust cues represented, (c) the sources of trust cues, and (d) the gender of the source?

2. Methodology

To answer the RQs, we conducted a quantitative content analysis of a sample of content from (digital) media sources most frequently used by public audiences in Germany to gain information about science (e.g. European Commission, 2021; Wissenschaft im Dialog, 2023).

Sample selection

We considered the heterogeneity of (digital) media environments and sought a sample encompassing outlets and accounts typically accessed by German publics when using media to acquire scientific information (e.g. European Commission, 2021). Next, we constructed four groups for comparison: (1) professional journalistic (i.e. print and online) media, including diverse outlets and formats that nevertheless all follow the same professional logic; (2) right-wing populist, non-mainstream media, which depart from journalism’s logic and are more likely to contain misleading and/or false information (e.g. Frischlich et al., 2022); (3) social media, for the purpose of better comparability, all with a reference to scientific accounts or contents; and (4) other Internet-based media (i.e. blogs and news aggregators), given their dissimilarity to the other groups. Table 1 provides an overview of the sample.

Table 1.

Sample overview.

Media sources Checked sample Study sample Number of trust cues per media source Average number of trust cues identified per media source
n % n % n % ø
Journalistic media 1374 75.8 664 73.3 4.635 78.1 6.98
Public television (ARD Tagesschau, ZDF heute, Quarks, Gut zu wissen) 30 1.66 12 1.32 66 1.1 5.50
Private television (RTL Aktuell, Sat.1 Nachrichten) 9 0.5 4 0.4 12 0.2 3.00
Print newspapers (Frankfurter Allgemeine Zeitung, Süddeutsche Zeitung, Spiegel, Zeit) 475 26.2 225 24.8 1461 24.6 6.49
Online newspapers (FAZ.net, SZ.de, spiegel.de, zeit.de) 666 36.8 324 35.8 2351 39.6 7.26
Tabloid newspaper (Bild, bild.de) 109 6.02 52 5.74 247 4.2 4.75
Popular special science magazines (Geo, P.M. Magazin, Spektrum der Wissenschaft) 85 4.7 47 5.2 496 8.4 10.55
Right-wing populist, non-mainstream media 72 4.0 36 4.0 266 4.5 7.38
Populist media (Epoch Times, Junge Freiheit, Compact) 72 4.0 36 4.0 266 4.5 7.38
Social media 123 6.8 67 7.4 139 2.3 2.07
Facebook (Wissenschaft aktuell, Harald Lesch Ultras, Fortschritt in der Wissenschaft) 57 3.1 32 3.5 68 1.1 2.13
Instagram (@doktorwissenschaft, @universumsfakten, @don.medicus, @diewissenschaftlerin) 6 0.3 4 0.4 5 0.1 1.25
X (c_drosten, dfg_public, BMBF_bund, helmholtz_de) 50 2.8 26 2.9 50 0.8 1.92
YouTube (MaiLab, Breaking Lab) 10 0.6 5 0.6 16 0.3 3.20
(Non-journalistic) internet-based media 243 13.4 139 15.3 892 15.0 6.43
Blogs (scienceblogs.de, scilogs.de) 27 1.5 15 1.7 61 1.0 4.07
News aggregators (t-online.de, web.de) 216 11.9 124 13.7 833 14.0 6.71
Total 1812 100 906 100 5932 100 6.55
Open in a new tab

Media content was collected for a full year, in seven constructed weeks from March 2022 to March 2023. Given the extensive array of media sources involved, we relied on multiple databases and methods for sample generation. 6 When possible, established search strings were used for the selection (Guenther et al., 2019; Schröder et al., 2025); 7 in cases that precluded using those search strings, we collected the content manually. The material was checked if it contained (1) an object of trust connected to science and (2) trust cues, and in total, n = 1812 were considered to be relevant. On that basis, we constructed a representative, randomized sample with half of those relevant pieces (n = 906).

Quantitative content analysis

In our quantitative content analysis, we followed a deductive approach with the help of a standardized codebook (see Document S1 in the Supplemental Material) developed based on inductive, qualitative content analysis in a previous study (Schröder et al., 2025). The codebook contains the 35 trust cues shown in Table 2. Because we wanted to compare various types of media, we focused exclusively on text and thus coded articles (i.e. print and online media), transcripts of videos (i.e. from TV and YouTube), and texts of individual posts (i.e. from social media). This poses a limitation that we will discuss later.

Four coders were trained and performed all coding, 8 which considered three formal criteria—media source (α = .99, CR = .99), media type (α = .99, CR = 1), and type of author (α = .80, CR = .91)—and four content-related criteria for each trust cue: 9 the source providing the respective trust cue (e.g. journalists, scientists, or social media users; α = .76, CR = .85), the source’s gender (α = .83, CR = .89), the level of each object of trust (i.e. micro, meso, or macro; α = .74, CR = .93), and the gender of micro-level objects of trust (α = .88, CR = .92).

To answer RQ1, we used descriptive statistics and chi-square tests. For RQ2, by contrast, we considered only micro-level codes that were clearly connected to female and male scientists (e.g. name mentioned or pronoun used). The categories “female,” “male,” and “other” were considered during the codebook’s development and the coding process; however, no coding for “other” was performed.

3. Results

Frequencies of trust cues (RQ1)

In total, we coded n = 5932 trust cues. Most trust cues were identified in journalistic media, followed by Internet-based media, populist media, and social media, and the cues most often referred to the dimension of expertise, followed by integrity, benevolence, transparency, and dialogue. Regarding the number of trust cues in relation to the amount of content from each type of media source, articles contained most trust cues in populist media, followed by journalistic media, further Internet-based media, and social media. However, a closer look revealed tremendous differences within the category of journalistic media; most trust cues were found in science magazines, followed by online and printed newspapers, public TV, and, trailing far behind, tabloid newspapers and private TV (see Table 1). The most common category of trust cues referred to qualification (i.e. dimension of expertise) and most often referred to the department or area of expertise and affiliation to an organization. Other categories of trust cues frequently referred to represented benevolence (i.e. social responsibility and benefits for society), integrity (i.e. scientific quality assurance and scientific standards and processes), and transparency (i.e. comprehensible language), as shown in Table 2.

Dimensions of trust in science (RQ1a)

The trust cues differed significantly across media types, χ2(12, n = 5932) = 83.361, p < .001, V = .068. In journalistic media, most trust cues found referred to expertise, followed by integrity, benevolence, transparency, and dialogue. This order is also true for social media, but with a higher frequency of integrity and benevolence cues, whereas transparency and dialogue cues appeared less often. Meanwhile, although expertise cues were also the most frequent in populist media, benevolence cues were the second-most common, followed by integrity cues and, further behind, dialogue and transparency cues. That order described further Internet-based media as well, in which benevolence cues were also the second-most frequent (see Table 3).

Table 3.

Frequencies of trust cues across (digital) media referring to dimensions of trust in science, different levels, and sources.

Journalistic media Populist media Internet-based media Social media Total
n % n % n % n % n %
Dimensions of trust in science*
Expertise 2148 46.3 115 43.2 355 39.8 42 30.2 2660 44.8
Integrity 1086 23.4 53 19.9 185 20.7 39 28.1 1363 23.0
Benevolence 1001 21.6 64 24.1 234 26.2 33 23.7 1332 22.5
Transparency 283 6.1 15 5.6 61 6.8 14 10.1 373 6.3
Dialogue 117 2.5 19 7.1 57 6.4 11 7.9 204 3.4
Levels of the object of trust in science*
Micro 3960 85.4 214 80.5 709 79.5 103 74.1 4986 84.1
Meso 626 13.5 51 19.2 174 19.5 30 21.6 881 14.9
Macro 49 1.1 1 0.4 9 1.0 6 4.3 65 1.1
Total* 4635 100 266 100 892 100 139 100 5932 100
Type of source**
Journalism 3539 80.6 171 75.7 605 75.9 / / 4315 77.7
Science 853 19.4 55 24.3 190 23.8 48 35.0 1146 20.6
Online actors / / / / 2 0.3 89 65.0 91 1.6
Total** 4392 100 226 100 797 100 137 100 5552 100
Open in a new tab
*

All codings were included in this calculation (n = 5932).**Only sources that could be assigned to a type of source were included (n = 5552).

Levels (RQ1b)

Significant differences additionally emerged between the use of trust cues connected to scientists (i.e. micro level), scientific organizations (i.e. meso level), 10 and the science system (i.e. macro level) across different (digital) media, χ2(6, n = 5932) = 46.113, p < .001, V = .062. Overall, trust cues mostly referred to science at the micro level, followed by science at the meso level and science at the macro level, meaning that the science system was rarely addressed. At the same time, proportionally more trust cues referred to the macro and the meso levels on social media despite occurring in very few cases (see Table 3). Micro-level objects occurred with the greatest frequency in journalistic media.

In general, although each dimension of trust was identified at all levels, not all trust cues were found at each level. At the macro level, trust cues referring to academic education and professional experience (i.e. expertise), participation in public events and media presence (i.e. dialogue), and the accessibility of results (i.e. transparency) were not identified. At the meso level, only trust cues referring to academic education (i.e. expertise) were absent, whereas all trust cues were identified at the micro level.

Sources of trust cues (RQ1c)

The sources of trust cues, namely journalism (e.g. journalists and news agencies), science (e.g. scientists and scientific organizations), and online actors, 11 differed across the media types, χ2(6, n = 5552) = 3570.726, p < .001, V = .567. Because journalistic content was the most prevalent type of media in the sample, journalism was unsurprisingly also the chief source of trust cues, followed by science and online actors. That ranking also held true when examining journalistic, populist, and other Internet-based media separately. On social media, online actors were the most common sources of trust cues, followed by science (see Table 3), and proportionally the greatest prevalence of trust cues from science was found on social media, followed by populist media. However, in absolute terms, science as a source was most common in journalism, thereby implying that source heterogeneity varied across (digital) media and, consequently, that trust cues may be used differently depending on the source of trust cues.

Gender of objects of trust (RQ2)

To answer RQ2, we considered only micro-level codings, 26% of which were for female scientists (n = 950) and 74% of which were for male scientists (n = 2684).

Dimensions of trust in science (RQ2a)

No significant difference emerged between the dimensions of trust in science addressed and the gender of the object of trust (see Table 4). That is, the most frequently addressed dimensions were the same for both genders: expertise, followed by benevolence, integrity, dialogue, and transparency.

Table 4.

Frequencies of dimensions of trust in science, categories and trust cues, sources of trust cues, and gender of the sources.

Female scientists Male scientists Total
n % n % n %
Dimensions of trust in science*
Expertise 546 57.5 1548 57.7 2094 57.6
Integrity 139 14.6 342 12.7 481 13.2
Benevolence 189 19.9 571 21.3 760 20.9
Transparency 30 3.2 110 4.1 140 3.9
Dialogue 46 4.8 113 4.2 159 4.4
Categories of trust cues*
Qualification (Expertise) 528 55.6 1488 55.4 2016 55.5
Social responsibility (Benevolence) 95 10.0 297 11.1 392 10.8
Benefits for society (Benevolence) 93 9.8 268 10.0 361 9.9
Scientific standards and processes (Integrity) 106 11.2 224 8.3 330 9.1
Trust cues*
Affiliation to an organization (Expertise) 231 24.3 596 22.2 827 22.8
Department or area/discipline of expertise (Expertise) 181 19.1 518 19.3 699 19.2
Assessment of public events/current affairs (Benevolence) 77 8.1 224 8.3 301 8.3
Professional position (Expertise) 56 5.9 196 7.3 252 6.9
Description (and explanation) of research processes 60 6.3 122 4.5 182 5.0
Total* 950 100 2684 100 3634 100
Sources of trust cues**
Journalism 693 76.6 1899 74.6 2592 75.1
Science 203 22.4 641 25.2 844 24.5
Online actors 9 1.0 6 0.2 15 0.4
Total** 905 100 2546 100 3451 100
Gender of sources***
Female sources 495 61.7 654 30.2 1149 38.8
Male sources 307 38.3 1509 69.8 1816 61.2
Total*** 802 100 2163 100 2965 100
Open in a new tab
*

All objects of trust in science at the micro level for whom gender was evident were included (n = 3634). **All sources at the micro level that could be assigned to journalism, science, or online actors and refer to female and male scientists were included (n = 3451). ***All sources and objects of trust in science at the micro level for whom gender was evident were included (n = 2965).

Trust cues (RQ2b)

Regarding trust cues and their superordinate categories, no significant gender differences were identified, meaning that female and male scientists were generally referred to in similar ways. Even so, examining the general distribution of trust cues could yield additional insights.

Overall, of all categories, 12 the following were most frequently mentioned for scientists: qualifications (i.e. expertise), social responsibility and benefits for society (i.e. benevolence), and scientific standards and processes (i.e. integrity).

Concerning trust cues, the most prevalent overall were organizational affiliation (i.e. expertise), department or area/discipline of expertise (i.e. expertise), assessment of public events or current affairs (i.e. benevolence), professional position (i.e. expertise), and descriptions (and explanations) of research processes (i.e. integrity). 13 On that basis, expertise seems to be the dimension of trust in science most often referred to, whereas transparency and dialogue seem to play only a minor role.

Sources of trust cues (RQ2c)

In general, all sources of trust cues referred to male scientists more often than female scientists in the context of trust in science. Furthermore, significant differences emerged regarding the sources of the cues and the gender of the object of trust, χ2(2, n = 3451) = 11.254, p < .01, V = .057: Journalistic media more often referred to male than female scientists, and those differences were similar for science as a source. However, this was not the case for online actors, who referred to female scientists more often than male scientists, even though there are fewer cases of online actors (see Table 4).

Gender of sources (RQ2d)

Significant differences also emerged between the gender of the source and the gender of the scientist referred to, χ2(1, n = 2965) = 244.357, p < .001, ϕ = .287. In the overall analysis (see Table 4), most trust cues that were clearly attributable to a source pertained to men, with significantly fewer directed toward women. Furthermore, whereas female sources of trust cues referred to both female and male scientists with almost equal frequency, for male sources of trust cues, the situation is different: they referred mostly to male scientists and only in a few cases to female scientists. From another angle, most trust cues used for female scientists originated from female sources of trust cues whereas most trust cues for male scientists came from male sources of trust cues.

4. Discussion

In our study, we investigated (digital) media as intermediaries of trust in science, with particular focus on the heterogeneity of digital media environments in terms of content, sources of trust cues, and objects of trust in science. In the process, we considered trust in science to be a multidimensional, multilevel construct and performed a quantitative content analysis of the most important sources that public audiences in Germany use to engage with science. Four of our findings merit sustained attention.

Key finding 1: Overall, expertise was the dimension of trust in science most frequently referred to, although differences across (digital) media also emerged

Our content analysis revealed that all dimensions of trust in science were addressed in all media types, with expertise cues being the most prevalent trust cues overall. As a consequence, exposure to expertise cues could exert the strongest trust-evoking impact on public audiences. At the same time, the dimensions of trust manifested to varying extents across diverse media types. Whereas social media placed greater emphasis on transparency and dialogue cues, populist and other Internet-based media prioritized benevolence as the second-most frequent dimension of trust in science. Thus, the media types emphasized different aspects, that is, dimensions of trust in content about science, and therefore media types are indeed heterogeneous in terms of their contents. That finding also implies varying effects of media types on audiences’ trust in science.

Key finding 2: Journalism is probably the most important source for trust cues

Considering the sources of trust cues that we investigated (i.e. journalism, science, and online actors), journalism contributed most to trust cues overall; science surfaced as the second-most frequent source, followed by online actors. That pattern generally held across journalistic, populist, and other Internet-based media; even so, on social media, online actors were the primary source, followed by science. Because most media content in our sample comes from journalistic sources, it is unsurprising that most sources also came from journalism. By extension, journalism could be the most influential source for audiences’ assessments of trust in science. The second-most common source was science because scientists were (in)directly quoted, provided insights, or even contributed entire pieces (e.g. in blogs) as a means to discuss scientific topics. This implies that communication by science itself plays an important role in evoking trust. Interestingly, science more often featured in populist media than other types of media. We did not assess tonality, and thus no inferences about pro- or antiscientific communication could be made from the data. However, in the coding process, coders had the impression that populist media are not inherently against science but instead have their own experts and are hostile only to certain scientific topics (e.g. gender, COVID-19, and climate change-related issues). Based on our results, social media seem to provide more heterogeneity in sources of trust cues than other types of media.

Key finding 3: Trust cues mostly referred to scientists

Concerning scientific objects of trust, trust cues predominantly pertained to science at the micro level, specifically to scientists. That finding suggests that tendencies toward personalization, as a noteworthy news factor in science media coverage (e.g. Amend and Secko, 2012; Guenther, 2019), could be transferred to other types of media as well. Nevertheless, on social media, trust cues more often related to the macro and meso levels than they did in other types of media. The increased emphasis on objects of trust in science at the meso level in social media may be attributed to the fact that organizations can communicate by using those platforms. However, the strong focus on the micro level in the sample prevented us from assessing whether different media types are heterogeneous in terms of objects of trust in science across the micro, meso, and macro levels. Nevertheless, individuals at the micro level potentially have the strongest impact on audiences’ trust, with possible spillover effects to the meso and macro levels.

Key finding 4: Concerning trust in science, our sample showed that female scientists are underrepresented, although Internet-based media have the potential to change that

Focusing on scientific objects of trust at the micro level, we did not identify any significant differences in the trust cues used and the dimensions of trust in science referred to between female and male scientists. Thus, female and male scientists were generally referred to in similar ways, even though how they were presented in terms of context and stereotypes may have varied (Schröder, 2025). Quantitatively speaking, however, women were underrepresented. Even so, such was not the case for online actors, who primarily referred to female scientists. Female sources of trust cues referenced both female and male scientists, albeit with men slightly dominating. The opposite, however, was true for male sources of trust cues, who referred to male scientists far more often than women. The gender of the sources of trust cues thus seems to have a major effect on the representation of female scientists in science media coverage, a finding that aligns with research showing the connection between the gender of the source and the scientists mentioned (e.g. GMMP, 2020; Kitzinger et al., 2008). However, findings with respect to the gender of the source can be partly explained by the fact that sources and objects of trust cues can be identical (e.g. a female scientist is quoted describing her academic education or a male scientist is quoted explaining his study). Nevertheless, journalism seems to mirror an imbalance in the scientific community, at least in Germany, namely that most individuals in the research landscape, especially in senior positions, are men (Nationale Akademie der Wissenschaften Leopoldina, 2022).

When linked to recent developments regarding public trust in science and its decline (e.g. Kennedy and Tyson, 2023; Weingart and Guenther, 2016), our findings show that different media refer to trust in science in different ways and that contents, sources of trust cues, and objects of trust in science may be more heterogeneous on social media and to some degree also in populist and Internet-based media than in journalistic media. Although those outcomes do not clearly indicate any potential positive or negative impacts of (digital) media on trust in science, our findings do reveal variations in which types of (digital) media mediate trust in science.

Based on our findings, future research should focus on how audiences perceive trust cues used in content about science and how such perceptions, along with differences between female and male scientific objects of trust, affect them. At the same time, audiences do not access information from only one medium but from a variety of media that together form their media repertoire. For that reason, only the composition of an individual’s media diet and the trust cues contained therein provide insights into the effects on trust in science. Future research should thus explore the extent to which exposure to trust cues in different media repertoires affects the stability or dynamics of trust in science within audiences (e.g. Guenther et al., 2024).

Beyond that, we identified trust cues solely at a textual level. To more precisely depict digital media and draw conclusions about the mediated environment of trust (in science), scholars need to conduct expanded content analyses that include multimodal elements such as (audio)visual and platform-specific cues that could serve as further heuristics (e.g. Metzger and Flanagin, 2013).

Limitations

Our findings come with limitations. The most crucial ones pertain to the sample, which was representative of the collected material but included only media mirroring the average media use by German publics (European Commission, 2021). Based on this media use, we chose exemplary media outlets and accounts and collected data from them during constructed weeks over the course of a year. Thus, only a selection of existing channels and accounts dealing with scientific content in social media were included. For that reason, although the data are representative of the overall data collected and of specific media outlets and accounts in our sample, they are not representative of media types in general (e.g. journalistic media in general).

In addition, regarding the content investigated, our analysis considered the textual level only, and the results pertain to science in general, not to specific scientific topics. For that reason, we could only approximate the content that audiences encounter about science in digital media environments.

Last, we considered trust cues to be trust-evoking and viewed trust and distrust as distinct concepts (e.g. Luhmann, 2014; Reif and Guenther, 2022; Resnick et al., 2015). Therefore, a separate analysis would be required to determine whether distrust can also be analyzed using those and/or different cues. Nonetheless, it is also possible that trust cues can be applied to alternative experts (e.g. in populist media).

Supplemental Material

sj-docx-1-pus-10.1177_09636625251337709 – Supplemental material for Mediating trust in content about science: Assessing trust cues in digital media environments
sj-docx-1-pus-10.1177_09636625251337709.docx (302.9KB, docx)

Supplemental material, sj-docx-1-pus-10.1177_09636625251337709 for Mediating trust in content about science: Assessing trust cues in digital media environments by Justin T. Schröder and Lars Guenther in Public Understanding of Science

Author biographies

Justin T. Schröder (MSc) is a research associate at LMU Munich’s Department of Media and Communication (IfKW), Germany, with a focus on science communication. In addition, he is perusing his PhD at the University of Hamburg, Germany.

Lars Guenther (PhD) is Professor of Communication Science at LMU Munich’s Department of Media and Communication (IfKW) in Germany, and Research Fellow at the Center for Research on Evaluation, Science and Technology (CREST) at Stellenbosch University in South Africa. He is interested into public perceptions of (controversial) science, science and health journalism, trust in science, as well as the public communication about risks and scientific (un)certainty.

1.

Most authors agree that trust and distrust are distinct concepts that do not merely represent two ends of the same scale (e.g. Luhmann, 2014; Reif and Guenther, 2022; Resnick et al., 2015). In this article, we focus exclusively on trust-evoking cues.

2.

Some studies have shown a positive relationship, particularly on social media (e.g. Huber et al., 2019), whereas others have reported a negative one, especially in the context of online media sources (e.g. Takahashi and Tandoc, 2016).

3.

Because (digital) media not only serve as objects of trust (i.e. trusted by subjects of trust) but also function as intermediaries (e.g. Reif and Guenther, 2022), two distinct research traditions have emerged: one focused on media’s credibility and views media as the object of trust, the other (i.e. media trust research) focused on media’s role as an intermediary of trust (e.g. Reif, 2021). In that context, our focus is science as the object and media as intermediaries of trust, which aligns with the tradition of media trust research.

4.

In the debate seeking to differentiate trust from trustworthiness, one proposal is that an object of trust can be perceived as being trustworthy, with trusting behavioral intentions defined as a willingness to make oneself vulnerable based on those perceptions (Mayer et al., 1995). Consequently, whereas the perception of science as the object of trust would exemplify trustworthiness, the behavior of a subject of trust based on such trustworthiness would exemplify trust. However, because trust cues focus on neither the subject nor object of trust but instead on information provided by intermediaries, neither the perception of the content nor the resulting behavior is investigated. Therefore, content-analytical trust research operates in the limbo between trustworthiness and trust. Because we understand such cues to be trust-evoking, we refer to trust cues instead of trustworthiness cues.

5.

Even so, men do outnumber women in fields of science, as is also true in Germany (e.g. Nationale Akademie der Wissenschaften Leopoldina, 2022). To some extent, media replicate that imbalance.

6.

We used MediathekView and OnlineTVrecorder for most TV newscasts and programs; Factiva for most print and online newspapers and magazines; FAZ Bibliotheksportal for Frankfurter Allgemeine Zeitung/FAZ.net; Google searches and manual savings for most of the populist media, blogs, and news aggregators; 4kdownloader for YouTube; manual savings for Facebook and Instagram posts; and TweetDownloader for Twitter (now X). Popular science magazines were purchased online.

7.

For print and online newspapers, populist media, blogs, and news aggregators, we used “wissenschaft*,” “studie*,” “forsch*,” “universität*,” and “institut*” (in English: “scien*,” “stud*,” “research*,” “universit*,” and “institut*”) to retrieve material likely to contain scientific content (see also Guenther et al., 2019).

8.

Four coders repeatedly coded the same media contents and subsequently discussed them. By way of those discussions, ambiguities were resolved, and the codebook was refined. Once a unified understanding was established, a pretest was conducted for reliability testing by coding 25 contributions and calculating reliability values. Weaknesses were examined more closely and discussed again. Last, a reliability test was conducted with 100 randomly selected pieces of media content.

9.

We used Krippendorff’s alpha (α) and Holsti (CR) to assess intercoder reliability.

10.

For the meso level, we further differentiated universities and university research (54%, n = 1603) from non-academic research institutions (18%, n = 537), government institutions (6%, n = 164), and private sector research (2%, n = 56). In total, 20% (n = 605) of all meso-level codings could not be clearly associated with one of those categories.

11.

Because online actors outside journalistic and scientific fields can be clearly distinguished from actors in those fields, they were grouped together despite being heterogeneous. Online actors included all individuals from social media and other Internet-based media, including interested groups of online users, bloggers, science influencers, science-related governmental organizations, and science funding organizations.

12.

Only categories ⩾5% in the overall sample are mentioned here.

13.

Only trust cues ⩾5% in the overall sample are mentioned here.

Footnotes

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research presented in this paper is part of the project “The Trust Relationship between Science and Digitized Publics” (TruSDi), funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—456602133. Grant applicants are Monika Taddicken (TA 712/4-1) and Lars Guenther (GU 1674/3-1). The project is coordinated by Anne Reif and supported by Peter Weingart in an advisory capacity. Further members of the research group are Justin T. Schröder, Evelyn Jonas, and Janise Brück.

Data availability statement: The data that support the findings of this study are available from the corresponding author [JTS], upon reasonable request.

Supplemental material: Supplemental material for this article is available online.

Contributor Information

Justin T. Schröder, Ludwig-Maximilians-Universität Munich, Germany; Universität Hamburg, Germany.

Lars Guenther, Ludwig-Maximilians-Universität Munich, Germany.

References

  1. Altenmüller MS, Nuding S, Gollwitzer M. (2021) No harm in being self-corrective: Self-criticism and reform intentions increase researchers’ epistemic trustworthiness and credibility in the eyes of the public. Public Understanding of Science 30(8): 962–976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Amend E, Secko DM. (2012) In the face of critique: A metasynthesis of the experiences of journalists covering health and science. Science Communication 34(2): 241–282. [Google Scholar]
  3. ARD/ZDF–Forschungskommission (2023) ARD/ZDF–Onlinestudie 2023. Available at: https://www.ard-zdf-onlinestudie.de/files/2023/ARD_ZDF_Onlinestudie_2023_Publikationscharts.pdf
  4. Bentele G. (1994) Öffentliches Vertrauen: Normative und soziale Grundlage für Public Relations [Public trust: Normative and social foundation for public relations]. In: Armbrecht W, Zabel U. (eds) Normative Aspekte der Public Relations: Grundlegende Fragen Und Perspektiven [Normative aspects of public relations: Fundamental questions and perspectives]. New York: Springer, pp. 131–158. [Google Scholar]
  5. Besley JC, Lee NM, Pressgrove G. (2020) Reassessing the variables used to measure public perceptions of scientists. Science Communication 43(1): 3–32. [Google Scholar]
  6. Brossard D. (2013) New media landscapes and the science information consumer. PNAS 110(3): 14096–14101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brossard D, Scheufele D. (2022) The chronic growing pains of communicating science online. Science 375(6581): 613–614. [DOI] [PubMed] [Google Scholar]
  8. Chimba M, Kitzinger J. (2009) Bimbo or boffin? Women in science: an analysis of media representations and how female scientists negotiate cultural contradictions. Public Understand of Science 19(5): 609–624. [DOI] [PubMed] [Google Scholar]
  9. European Commission (2021) Eurobarometer: European Citizens’ Knowledge and Attitudes Towards Science and Technology. Available at: https://europa.eu/eurobarometer/surveys/detail/2237
  10. Fletcher S, Joe MB, Hernandez S, Toman I, Harrison TG, Ruzycki SM. (2021) The gender of COVID-19 experts in newspaper articles: A descriptive cross-sectional study. Journal of General Internal Medicine 36(4): 1011–1016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Frischlich L, Kuhfeldt L, Schatto-Eckrodt T, Clever L. (2022) Alternative counter-news use and fake news recall during the COVID-19 crisis. Digital Journalism 11(1): 80–102. [Google Scholar]
  12. Giddens A. (1990) The Consequences of Modernity. Cambridge: Polity Press. [Google Scholar]
  13. GMMP (2020) Global media monitoring project 2020. Available at: https://whomakesthenews.org/wp-content/uploads/2021/11/GMMP2020.ENG_.FINAL_.pdf
  14. Goldenberg MJ. (2023) Public trust in science. Interdisciplinary Science Reviews 48(2): 366–378. [Google Scholar]
  15. Guenther L. (2019) Science Journalism. Oxford: Oxford University Press. [Google Scholar]
  16. Guenther L, Bischoff J, Löwe A, Marzinkowski H, Voigt M. (2019) Scientific evidence and science journalism: Analysing the representation of (un)certainty in German print and online media. Journalism Studies 20(1): 40–59. [Google Scholar]
  17. Guenther L, Schröder JT, Reif A, Brück J, Taddicken M, Weingart P, et al. (2024) Intermediaries in the limelight: How exposure to trust cues in content about science affects public trust in science. Journal of Science Communication 23(09): A06. [Google Scholar]
  18. Hendriks F, Kienhues D, Bromme R. (2015) Measuring laypeople’s trust in experts in a digital age: The Muenster Epistemic Trustworthiness Inventory (METI). PLoS ONE 10(10): e0139309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hendriks F, Kienhues D, Bromme R. (2016) Evoking vigilance. Would you (dis)trust a scientist who discusses ethical implications of research in a science blog? Public Understanding of Science 25(8): 992–1008. [DOI] [PubMed] [Google Scholar]
  20. Hijmans E, Pleijter A, Wester F. (2003) Covering scientific research in dutch newspapers. Science Communication 25(2): 153–176. [Google Scholar]
  21. Hofstra B, Kulkarni VV, Galvez SM-N, Jurafsky D, McFarland DA. (2020) The diversity–innovation paradox in science. PNAS 117(17): 9284–9291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Huber B, Barnidge M, Gil de, Zúñiga H, Liu J. (2019) Fostering public trust in science: The role of social media. Public Understanding of Science 28(7): 759–777. [DOI] [PubMed] [Google Scholar]
  23. Hubner AY, Bullock OM. (2024) Why science should have a female face: Female experts increase liking, competence, and trust in science. Science Communication. Epub ahead of print 12 November. DOI: 10.1177/10755470241295676. [DOI] [Google Scholar]
  24. Jennings W, Stoker G, Bunting H, Valgarðsson VO, Gaskell J, Devine D, et al. (2021) Lack of trust, conspiracy beliefs, and social media use predict COVID-19 vaccine hesitancy. Vaccines 9(6): 906059. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Joubert M, Guenther L, Rademan L. (2022) Expert voices in South African mass media during the COVID-19 pandemic. South African Journal of Science 118(5/6): 12480. [Google Scholar]
  26. Kennedy B, Tyson A. (2023) Americans’ trust in scientists, positive views of science continue to decline. Pew Research Center. Available at: https://www.pewresearch.org/science/2023/11/14/americans-trust-in-scientists-positive-views-of-science-continue-to-decline/
  27. Kitzinger J, Chimba M, Williams A, Haran J, Boyce T. (2008) Gender, stereotypes and expertise in the press: How newspapers represent female and male scientists. Cardiff School of Journalism. Available at: https://orca.cf.ac.uk/28633/1/Kitzinger_Report_2.pdf
  28. Kohring M. (2016) Misunderstanding trust in science: A critique of the traditional Discourse on science communication. Journal of Science Communication 15(5): C04. [Google Scholar]
  29. Luhmann N. (2014) Vertrauen: Ein Mechanismus der Reduktion sozialer Komplexität [Trust: A mechanism for reducing social complexity]. Konstanz: UVK. [Google Scholar]
  30. Mayer RC, Davis JH, Schoorman FD. (1995) An integrative model of organizational trust. The Academy of Management Review 20(3): 709–734. [Google Scholar]
  31. Metzger MJ, Flanagin AJ. (2013) Credibility and trust of information in online environments: The use of cognitive heuristics. Journal of Pragmatics 59(PtB): 210–220. [Google Scholar]
  32. Mitchell M, McKinnon M. (2019) “Human” or “objective” faces of science? Gender stereotypes and the representation of scientists in the media. Public Understanding of Science 28(2): 177–190. [DOI] [PubMed] [Google Scholar]
  33. Nationale Akademie der Wissenschaften Leopoldina [National Academy of Sciences Leopoldina] (2022) Frauen in der Wissenschaft: Entwicklungen und Empfehlungen [Women in Science: Developments and Recommendations]. Available at: 2022_Leopoldina_Stellungnahme_Frauen-in-der-Wissenschaft_Web.pdf
  34. Niemi MK, Pitkänen V. (2017) Gendered use of experts in the media: Analysis of the gender gap in Finnish news journalism. Public Understanding of Science 26(3): 355–368. [DOI] [PubMed] [Google Scholar]
  35. Origgi G. (2012) Epistemic injustice and epistemic trust. Social Epistemology 26(2): 221–235. [Google Scholar]
  36. Pavlik J. (2000) The impact of technology on journalism. Journalism Studies 1(2): 229–237. [Google Scholar]
  37. Plohl N, Musil B. (2021) Modeling compliance with COVID-19 prevention guidelines: The critical role of trust in science. Psychology, Health & Medicine 26(1): 1–12. [DOI] [PubMed] [Google Scholar]
  38. Ratcliff CL, Wicke R. (2023) How the public evaluates media representations of uncertain science: An integrated explanatory framework. Public Understanding of Science 32(4): 410–427. [DOI] [PubMed] [Google Scholar]
  39. Reich Z. (2011) Source credibility as a journalistic work tool. In:Franklin B, Carlson M. (eds) Journalists, Sources, and Credibility: New Perspectives. London: Routledge, pp. 19–36. [Google Scholar]
  40. Reif A. (2021) Mehr Raum für Vertrauen? Potenzielle Veränderungen des Vertrauens in Wissenschaft durch partizipative Online-Umgebungen [More room for trust? Potential changes in trust in science through participatory online environments]. In: Döbler T, Pentzold C, Katzenbach C. (eds) Räume Digitaler Kommunikation: Lokalität—Imagination–Visualisierung [Spaces of digital communication: Locality—Imagination—Visualization]. Herbert von Halem, pp. 210–243. [Google Scholar]
  41. Reif A, Guenther L. (2022) How representative surveys measure public (dis)trust in science: A systematisation and analysis of survey items and open-ended questions. Journal of Trust Research 11(2): 94–118. [Google Scholar]
  42. Reif A, Taddicken M, Guenther L, Schröder JT, Weingart P. (2024) The public trust in science scale: A multilevel and multidimensional approach. Science Communication. Epub ahead of print 20 December. DOI: 10.1177/10755470241302758. [DOI] [Google Scholar]
  43. Resnick HE, Huddleston N, Sawyer K. (2015) Trust and Confidence at the Interfaces of the Life Sciences and Society: Does the Public Trust Science? A Workshop Summary. Washington, DC: The National Academies Press. [PubMed] [Google Scholar]
  44. Rosman T, Bosnjak M, Silber H, Koßmann J, Heycke T. (2022) Open science and public trust in science: Results from two studies. Public Understanding of Science 31(8): 1046–1062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Saffran L, Hu S, Hinnant A, Scherer LD, Nagel SC. (2020) Constructing and influencing perceived authenticity in science communication: Experimenting with narrative. PLoS ONE 15(1): e0226711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schäfer MS. (2016) Mediated trust in science: concept, measurement and perspectives for the “science of science communication..” Journal of Science Communication 15(5): C02. [Google Scholar]
  47. Schröder JT, Guenther L. (2024) Science and health journalism in the digital age and its implications on public trust. In: Walsh-Childers K, McKinnon M. (eds) The Palgrave Handbook of Health and Science Journalism. Basingstoke: Palgrave Macmillan, pp.421–438. [Google Scholar]
  48. Schröder JT, Brück J, Guenther L. (2025) Identifying trust cues: how trust in science is mediated in content about science. Journal of Science Communication 24(01): A06. [Google Scholar]
  49. Schröder JT. (2025) Trust cues in content about science: How the media presents female and male scientists differently. In: Fage-Butler A, Ledderer L, Nielsen KH. (eds) Science Communication and Trust. London: Palgrave Macmillan, pp. 385–406. [Google Scholar]
  50. Song H, Markowitz DM, Taylor SH. (2022) Trusting on the shoulders of open giants? Open science increases trust in science for the public and academics. Journal of Communication 72(4): 497–510. [Google Scholar]
  51. Sperber D, Clément F, Heintz C, Mascaro O, Mercier H, Origgi G, et al. (2010) Epistemic vigilance. Mind & Language 25(4): 359–393. [Google Scholar]
  52. Taddicken M, Krämer N. (2021) Public online engagement with science information: On the road to a theoretical framework and a future research agenda. Journal of Science Communication 20(3): A05. [Google Scholar]
  53. Takahashi B, Tandoc EC. (2016) Media sources, credibility, and perceptions of science: Learning about how people learn about science. Public Understanding of Science 25(6): 674–690. [DOI] [PubMed] [Google Scholar]
  54. Weingart P. (2017) Wissenschaftskommunikation unter digitalen Bedingungen. Funktionen, Akteure und Probleme des Vertrauens [Science communication under digital conditions: Functions, actors, and problems of trust]. In: Weingart P, Wormer H, Wenninger A, Hüttl RF. (eds) Perspektiven der Wissenschaftskommunikation im Digitalen Zeitalter [Perspectives on science communication in the digital age]. Weilerswist-Metternich: Velbrück Wissenschaft, pp. 31–59. [Google Scholar]
  55. Weingart P, Guenther L. (2016) Science communication and the issue of trust. Journal of Science Communication 15(5): 1–11. [Google Scholar]
  56. Welzenbach-Vogel IC, Milde J, Stengel K, Dern M. (2021) Vertrauen in das Gelingen der Energiewende durch Medienberichterstattung? Eine Inhaltsanalyse von journalistischen Beiträgen auf deutschen Online-Nachrichtenportalen unter Berücksichtigung von vertrauensrelevanten Aussagen zu an der Energiewende beteiligten Akteuren [Trust in the success of the energy transition through media coverage? A content analysis of journalistic articles on German online news portals with a focus on trust-relevant statements about actors involved in the energy transition]. In: Milde J, Welzenbach-Vogel IC, Dern M. (eds) Intention und Rezeption von Wissenschaftskommunikation [Intention and reception of science communication]. Cologne: Herbert von Halem Verlag, pp. 65–86. [Google Scholar]
  57. Wissenschaft im Dialog [Science in Dialogue] (2023) Wissenschaftsbarometer 2023 [Science Barometer Germany 2023]. Available at: https://wissenschaft-im-dialog.de/projekte/wissenschaftsbarometer/#erhebung-2023
  58. Xiao X, Borah P, Su Y. (2021) The dangers of blind trust: Examining the interplay among social media news use, misinformation identification, and news trust on conspiracy beliefs. Public Understanding of Science 30(8): 977–992. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

sj-docx-1-pus-10.1177_09636625251337709 – Supplemental material for Mediating trust in content about science: Assessing trust cues in digital media environments
sj-docx-1-pus-10.1177_09636625251337709.docx (302.9KB, docx)

Supplemental material, sj-docx-1-pus-10.1177_09636625251337709 for Mediating trust in content about science: Assessing trust cues in digital media environments by Justin T. Schröder and Lars Guenther in Public Understanding of Science

Articles from Public Understanding of Science (Bristol, England) are provided here courtesy of SAGE Publications

RESOURCES