Uncivil communication and epistemic trustworthiness concerns in public online discussions in response to scientists during the Covid-19 pandemic

1 Introduction

The Covid-19 pandemic, while being a matter of significant societal concern, also exemplifies what is known as a ‘socio-scientific issue’ [Sadler, Barab & Scott, 2007]. Such issues are characterized by their complex interplay between scientific knowledge and societal values. Socio-scientific issues cannot be resolved by scientific practices alone; they require solutions that integrate moral, political, social, economic, and ethical considerations [Sadler et al., 2007]. Throughout the pandemic, social media platforms such as Twitter/X¹ served as public arenas where practical solutions and epistemic knowledge were negotiated. This dynamic process was particularly pronounced due to the continuous influx of newly published scientific evidence, occurring almost daily. Discussions around the pandemic were not only pursued by the general public but also by scientists. In Germany, some scientists used Twitter/X to disseminate findings and to address politics by making political calls for action or evaluating measures [Biermann, Peters & Taddicken, 2023]. In particular, some virologists gained popularity that was also reflected in their number of Twitter/X followers [Utz, Gaiser & Wolfers, 2022]. However, increased visibility led to scientists facing harassment and attacks on their credibility [Nogrady, 2021]. This underscores the relevance of researching direct science communication in online environments.

The possibilities of direct online communication from scientists to the public and vice versa create a dynamic interplay with potential advantages and inherent challenges. This study addresses two key challenges: the trust relationship between science and the public and deliberative challenges arising from incivility. Scientific information is usually complex and intrinsically uncertain [Popper, 2003]. Due to a limited comprehension of scientific information by non-scientists, reliance on and trust in science is vital, especially within global challenges and crises such as the pandemic where adherence to protection measures depends on trust in science [Dohle, Wingen & Schreiber, 2020; Wintterlin et al., 2022]. Furthermore, deliberative challenges are brought to the forefront. During the pandemic, responses to scientists’ tweets sometimes veered into uncivil communication, with scientists even reporting threats after communicating on social media [Nogrady, 2021]. For a public negotiation of issue solutions, uncivil communication can have negative consequences as it impedes the democratic goal of deliberation [Anderson, Brossard, Scheufele, Xenos & Ladwig, 2013; Papacharissi, 2004]. Uncivil communication comprises derogatory forms of communication that violate social discourse norms or democratic principles. By conducting manual content analysis, we examine the prevalence of uncivil communication and concerns about epistemic trustworthiness directed at scientists on Twitter/X. Through this investigation, the aim is to gain a more detailed understanding of the dynamics of uncivil communication in the context of online science communication, particularly around socio-scientific issues. It is unclear whether such uncivil communication also impacts the trustworthiness of participating actors. Thus, a potential relationship between the co-occurrence of incivility and trustworthiness concerns is investigated. Therefore, this study provides insights into the dynamics of online discussions around a socio-scientific issue by investigating the interdependence of two challenges scientists face when communicating on social media: uncivil communication and trustworthiness concerns.

2 Background

2.1 Science communication on social media during the Covid-19 pandemic

The public relied on various information and evidence types to cope with the pandemic [Lu, Chu & Ma, 2021]. While social media provides direct access to unlimited content, research indicates that inaccurate information and fake news are disseminated more easily than factual information [Cinelli et al., 2020; Vosoughi, Roy & Aral, 2018]. Moreover, social media can foster inaccurate science communication and amplify science-skeptical views [Anderson & Huntington, 2017]. However, social media was a popular online source for pandemic-related information [Wissenschaft im Dialog, 2023]. New visible scientists emerged during the pandemic, communicating directly to the public. The term “visible scientist” originally referred to prominent scientists in mass media who brought issues to the media agenda [Goodell, 1977]. Today, there are various possibilities for scientists to acquire visibility, inter alia, through social media [Biermann et al., 2023; Olesk, 2021]. This direct online communication of scientists toward the public has the potential to educate the public and correct misinformation by writing comments and actively engaging in public discourse [Taddicken & Krämer, 2021]. In Germany, several scientists made use of these potentials and directly communicated toward a broader public or toward politicians on social media, such as Twitter/X [Drescher et al., 2021].

While scientific evidence on the transmission of the virus was scarce at the beginning of the pandemic, new evidence appeared almost daily and was often communicated publicly at an early stage. This led to a public negotiation of epistemic knowledge on social media. It sometimes preceded the scholarly discourse or happened concurrently, providing the broader public with opportunities to raise trustworthiness concerns while observing scientific disagreement [Broer & Hasebrink, 2022]. Therefore, the pandemic may represent the transformation of the knowledge system where roles, phases, contexts, and hierarchies become flexible due to a change in values, digitalization processes, and conflicting knowledge claims [Neuberger et al., 2023]. A strict differentiation between scholarly discourse, mass-media communication, and dialogue-oriented discourse becomes increasingly challenging as diverse actors negotiate epistemic knowledge [Broer & Hasebrink, 2022]. Thus, the intended recipient or audience of public social media communication may not always be definite and deviate from the imagined audience of the communicator, promoting the risk of context collapse, such as the flattening of multiple audiences into one [Marwick & boyd, 2010]. When communicating online, scientists need to be aware of the context in which they are communicating [Martini, Battisti, Bina & Consolandi, 2022].

2.2 Uncivil communication in online discussions

A free and respectful exchange of ideas is considered a democratic goal. This approach to democracy lies behind the concept of deliberation and aims to identify solutions by exchanging arguments publicly and involving citizens in the discussions [Habermas, 1985]. However, while public discourse has always been partly uncivil, in online environments, uncivil communication is able to spread faster and wider than before [Coe, Kenski & Rains, 2014]. Moreover, due to mainly text-based communication, the absence of face-to-face communication in an online context can foster more heated discussions [Papacharissi, 2004] and communication is often more informal and personalized [Ward & McLoughlin, 2020]. Moreover, emotional, heated discussions sometimes take an uncivil route [Anderson et al., 2013].

Incivility as a conceptual framework to investigate debate is rooted in political science and political communication research, but it is not only suitable for political debate. Besides, there is a difference of opinion over what incivility and uncivil communication comprise. Papacharissi [2004] defines incivility as “disrespect for the collective traditions of democracy” [p. 267], whereas others see it as explicit attacks that insult another person’s character and that detract from healthy debate [Anderson & Huntington, 2017; Brooks & Geer, 2007]. Also, while some conceptual definitions differentiate between incivility and impoliteness — with impoliteness comprising lighter forms such as vulgarity, sarcasm, or the usage of all-caps to imply shouting — others consider incivility as a subcategory of impoliteness, which also refers to moral issues or democracy [Papacharissi, 2004; Theocharis, Barberá, Fazekas, Popa & Parnet, 2016]. The contestation around incivility’s understanding also lies in subjectivity. What one person considers uncivil might be appropriate and civil for another. Incivility may thus be best described as a continuum [Sydnor, 2017]. However, there is a concurring understanding that incivility is a violation of norms, but there is not yet an understanding as to which norm violations are considered uncivil [Bormann, Tranow, Vowe & Ziegele, 2021]. Thus, some studies include public perceptions of norm violations and sanction worthiness to classify violations as uncivil [e.g., Bormann et al., 2021; Muddiman, 2017].

Even though incivility may not be a clear term and its presence cannot be reliably compared, the effects of uncivil communication on engagement behavior, opinion formation, and issue perceptions [e.g., Anderson et al., 2013; Borah, 2012; Chinn & Hart, 2021] show the necessity to further explore this phenomenon, and to sharpen the conceptualization, especially around an issue with high societal relevance and many controversies such as the Covid-19 pandemic. Researchers have discovered that exposure to incivility can polarize risk perception on a scientific issue [Anderson et al., 2013]. Incivility can also increase aggressiveness [Gervais, 2015; Rösner, Winter & Krämer, 2016] or lead people to retaliate [Masullo Chen & Lu, 2017; Masullo Chen, Muddiman, Wilner, Pariser & Stroud, 2019]. Therefore, aggressive language can affect the trustworthiness the person using it, negatively impact the credibility of the information, and may lead to the impression that less was learned [König & Jucks, 2019]. Moreover, being exposed to uncivil communication can also increase the use of incivility by the recipient and increase intentions to participate politically [Masullo Chen et al., 2019]. Within science communication, (uncivil) disagreement amongst scientists also has negative effects. Civil and uncivil disagreement online between scientists can lead to lower acceptance of the scientific issue, as well as lower trust in science, scientific methods and less attention towards the issue at hand [Chinn & Hart, 2021]. Besides the effects of uncivil debate on recipients, uncivil disagreement can also affect the target or the initial communicator — in this case, the scientist. Personal attacks may decrease scientists’ willingness to attend future media interviews [Nogrady, 2021]. However, it may also be outweighed by a motivation to provide expertise to the public discourse, especially in a crisis such as the Covid-19 pandemic [Nölleke, Leonhardt & Hanusch, 2023].

In summary, throughout this paper, the terms “incivility” and “uncivil communication” synonymously refer to forms of communication that violate social discourse norms and/or democratic principles. This includes language that implies shouting, using vulgarity or sarcasm, name-calling, threats, silencing, and sexist or stereotypical remarks. Therefore, this uncivil communication may have destructive consequences when negotiating socio-scientific issues.

2.3 Raising epistemic trustworthiness concerns online

During crises such as the Covid-19 pandemic, reliance on and trust in scientific knowledge is vital as protection measures also depend on trust in science [Wintterlin et al., 2022]. Public trust generally refers to the complex concept defined as a mechanism for reducing complexity and coping with uncertainties involving the object and subject of trust, i.e., the recipient and the giver of trust [Reif & Guenther, 2021; Luhmann, 2014; Giddens, 2007]. Direct communication by scientists offers the public the chance to develop opinions regarding the trustworthiness of scientists [Wintterlin et al., 2022]. In general, the potential for direct online participation in science can strengthen the trust relationship between science and the public [Reif, 2021]. In Germany, trust in science initially experienced a substantial increase at the onset of the pandemic and subsequently maintained a high level. This high level of trust has slightly decreased as the pandemic has progressed maintaining a high level [Bromme, Mede, Thomm, Kremer & Ziegler, 2022]. However, concerning the issue of Covid-19, some polarization existed in terms of skeptical beliefs about the evidence related to Covid-19 [Bromme et al., 2022]. In direct exchanges with scientists, decreasing trust and growing polarizing attitudes in terms of science rejection might have become visible, and discussants might question the epistemic trustworthiness of scientists. Epistemic trustworthiness describes experts’ characteristics that determine whether individuals will rely on or acknowledge them regarding scientific information because of laypeople’s limited knowledge in this field [Hendriks, Kienhues & Bromme, 2015]. Epistemic trustworthiness is a multidimensional concept, with crucial factors considered as expertise, integrity, and benevolence [Hendriks et al., 2015; Mayer, Davis & Schoorman, 1995].

Examining publicly expressed concerns about the trustworthiness of scientists is relevant insofar as it may have further effects. It may influence how recipients assess the credibility of the claim, or how they estimate the trustworthiness of the source, i.e., the scientist [Gierth & Bromme, 2019]. In controversial science debates, trust in science is often the main focus [Laslo & Baram-Tsabari, 2019]. So far, questioning or doubting epistemic trustworthiness has been investigated as, inter alia, science-critical user comments [Gierth & Bromme, 2019], as ad hominem attacks [Barnes, Johnston, MacKenzie, Tobin & Taglang, 2018], or as ethical concerns toward science [Laslo & Baram-Tsabari, 2019]. While these are different phenomena, attitudes toward science and scientists may be affected. However, what has to be considered is that concerns referring to scientists’ expertise, integrity, or benevolence can also be very constructive as critique is a usual and necessary dimension of democracy [Egelhofer, 2023]. Moreover, within scholarly discourse, claims are often scientifically criticized, which is part of the scholarly discourse and ensures the quality of research. Scientists’ disagreement can be considered a normal process to identify knowledge gaps [Dieckmann et al., 2015]. When scientific disagreements are discussed publicly online, they can have negative consequences as these might disrupt the picture of science as a provider of reliable evidence to decide on political measures [Broer & Hasebrink, 2022]. Scientific disagreement may also influence the public’s ascription of competence toward scientists [Dieckmann et al., 2015].

2.4 Examining uncivil communication and trustworthiness concerns in socio-scientific debates

Within uncertain socio-scientific issues, debates can become controversial and heated due to their high societal relevance and urgency. Identifying solutions for such challenges requires a trusting relationship between science and society. Additionally, deliberate and civil engagement from actors across various fields is vital, as solutions to socio-scientific issues cannot rely solely on scientific practices but must also consider moral and ethical aspects [Sadler et al., 2007].

This study examines incivility and trustworthiness concerns as distinct concepts. Trustworthiness concerns can be expressed civilly, and raising justified concerns about the trustworthiness of scientists can be constructive. Conversely, unfounded trustworthiness concerns can undermine deliberation [e.g. Gierth & Bromme, 2019] and thus may be perceived as a form of uncivil communication within socio-scientific debate. Despite this, the study does not classify unfounded trustworthiness concerns as a form of incivility, as incivility as a concept is rooted within the context of political debate. Epistemic trustworthiness concerns, on the other hand, specifically pertain to scientific actors. Furthermore, whether these concerns are justified or not is often subjective and not easily assessable from an external perspective. Therefore, this study investigates incivility and epistemic trustworthiness concerns as distinct constructs while not differentiating between unfounded and justified trustworthiness concerns. The co-occurrences of incivility and trustworthiness concerns are examined to provide insights into how these two forms of disruptions exist within the negotiation of a socio-scientific issue, the Covid-19 pandemic, thus shedding light on their dynamics.

3 Research questions

In order to gain insights into the dynamics of uncivil communication and concerns about the epistemic trustworthiness, this study investigates the prevalence of incivility forms and trustworthiness concerns within Twitter/X replies towards visible German virologists. Against the background of previous research, which has shown that uncivil communication is a common feature of online discussions, studies have also concluded that uncivil comments account for a minority of overall comments. Research investigating incivility in various online communication contexts suggests that most tweets do not contain uncivil communication [e.g., Anderson & Huntington, 2017; Coe et al., 2014; Collins & Nerlich, 2014; Papacharissi, 2004]. Exact shares of incivility vary, which might be explained through different measurements but is also rooted in the topic of discussion and the platform on which the discussion takes place [e.g., Coe et al., 2014; Su et al., 2018; Theocharis, Barberá, Fazekas & Popa, 2020]. As definitions and operationalizations of incivility are contested, this study investigates individual forms of uncivil communication separately. Thus, it is easier to compare results with other conceptualizations and operationalizations. Content analyses that investigate individual forms of uncivil communication are scarce, and they are rarely conducted in the context of science communication or science-related content. Consequently, this study aims to answer the following research question:

RQ₁: To what extent are replies toward visible German virologists’ tweets uncivil, and what forms of incivility are predominantly used?

Within its first year, the Covid-19 pandemic overshadowed almost every other issue on the political and news agenda [Rauchfleisch, Siegen & Vogler, 2021]. During this time, there was conflicting and uncertain information regarding the virus’s spread, vaccination progress and safety, and the necessity of lockdown or containment measures. Over time, certain subjects, particularly those revolving around political responses to the pandemic, became increasingly polarized, as evidenced in debates on topics such as mask-wearing [Lang, Erickson & Jing-Schmidt, 2021; Pascual-Ferrá, Alperstein, Barnett & Rimal, 2021] and vaccination [Jiang et al., 2021]. Previous research has found that discussion associated with polarized issues, tend to be uncivil [Anderson et al., 2013]. As the aim of this study is to ascertain the dynamics of incivility by relating the occurrences of incivility to the co-occurrences of incivility and concerns raised about epistemic trustworthiness concerns, the development of tweets over time containing these concerns is also investigated.

RQ₂: How do forms of incivility in replies toward visible German virologists’ tweets and concerns about epistemic trustworthiness develop over the course of the pandemic?

A previous study by Anderson and Huntington [2017] shows that uncivil communication often co-occurs with climate-skeptical perspectives within discussions around extreme weather events. Social media might also fuel science-skeptical perspectives toward science, which has already become apparent for issues such as vaccines [Kata, 2012] and climate change [Anderson & Huntington, 2017]. Direct interactions between scientists and the public on Twitter/X offer the potential for users to communicate trust attitudes toward scientists directly. Uncivil communication and science-skeptical attitudes can negatively affect recipients’ attitudes toward communicators of the debates and the debated issues [Anderson et al., 2013; Chinn & Hart, 2021]. It can be assumed that the co-occurrence of incivility and concerns about trustworthiness can exacerbate negative effects. Moreover, exposure to uncivil communication may motivate users to raise trustworthiness concerns about the initial communicator. Also, in a controversial debate, raised trustworthiness concerns can foster more uncivil communication and further polarize the discussion. Thus, the interplay of these two phenomena is investigated.

RQ₃: Does the occurrence of uncivil communication relate to the occurrence of concerns about the epistemic trustworthiness of scientists?

4 Methods

To answer the research questions, a manual content analysis of Twitter/X replies (N = 6,000) directed toward six visible German virologists was conducted by two coders. This study focuses on the replies toward scientists in the field of virology, as they were the experts in the field that gained the most attention, in particular during the beginning of the pandemic. The selected scientists are all professors of virology and are actively engaged in research. At the time of data collection, they had over 10,000 followers on Twitter/X. The selected sample consisted of three male and three female virologists, thus minimizing potential gender effects.

This study’s sample encompasses reply tweets from four distinct periods within the German pandemic context: the onset of the first lockdown (16.03.2020–28.03.2020), the onset of the second lockdown (28.10.2020–10.11.2020), the middle phase of the second lockdown (16.12.2020–03.01.2021), which encompassed the Christmas holidays and the beginning of the vaccine rollout, and lastly, the concluding phase of the second lockdown (03.03.2021–16.03.2021). Data for the identified scientists and time periods were tracked by a crawler that made use of Twitter’s open-source library. The crawler was based on the social media analytics framework [Stieglitz, Mirbabaie, Ross & Neuberger, 2018]. A stratified sampling approach was employed to select the reply tweets. To ensure proportional representation from each time period, the data was divided into strata based on the time periods. Subsequently, 6,000 reply tweets were sampled from each stratum proportionately to the subsample size, thus accounting for temporal variations (t₁: n = 299; t₂: n = 2017; t₃: n = 2995; t₄: n = 689). The number of reply tweets during these time frames varies quite strongly. The possible reasons for this may include the number of original tweets by scientists, specific (political) events (e.g., changes in lockdown measures), or certain statements or actions by scientists or politicians that garnered significant attention and encouraged dialogue. A proportionately weighted sample achieves a more accurate representation of the dynamics in each time period. Changes in attitudes toward science over time might become visible, in particular during ongoing lockdown measures, possibly in combination with strong socializing restrictions over Christmas and New Year celebrations, the increasing controversialities and polarization on political measures, and the beginning of the first vaccine rollout.

The codebook (see supplementary material) that formed the basis for this analysis included formal categories such as the time period, gender of the virologist, and whether the tweet contained a personal address (e.g., "You are a…" or "You should make…") or not (e.g., "They should all…" or "He is…"). We analyzed whether the tweets contained any form of uncivil communication, regardless of who it was directed at. If they did contain uncivil communication, we further scrutinized the manifestations of incivility present, derived from former conceptualizations and operationalizations from the literature [e.g., Anderson & Huntington, 2017; Papacharissi, 2004; Southern & Harmer, 2019; Theocharis et al., 2016]. The forms of uncivil communication under investigation encompassed emotionalizing language (including the use of all-caps or overuse of punctuation that implied shouting, e.g., “You are SO RACIST!!!!!”), sarcasm/cynical language (e.g., “This hurts! Is she talking to fourth graders or adults?”), vulgarity (e.g., “Germany is testing shit and above all wrong!”), name-calling (e.g., “You are a liar and scaremonger”), violence and threats (e.g., “You are sussed, the clock is ticking”), sexist or stereotypical remarks (e.g., “Snooty, know-it-all elite societies don’t have long lives”), and silencing (e.g., “anyone who […] is always wrong and still shoots his mouth off, simply has no right to do so anymore”). Sexist and stereotypical remarks were combined in one variable as sexist remarks often included gender stereotypes, and a distinction between these two variables was not always clear.

Subsequently, we examined whether the reply tweet raised trustworthiness concerns, focusing on epistemic trustworthiness dimensions expertise, integrity, and benevolence derived from the Muenster Epistemic Trustworthiness Inventory (METI) [Hendriks et al., 2015]. We examined whether users questioned, doubted, or denied the epistemic trustworthiness of the scientists, i.e., their expertise (e.g., “For me, you are a conspiracy theorist with no in-depth knowledge”), integrity (e.g., “Stop the lies. Pull yourself together at last!”) or benevolence (e.g., “It is irresponsible how she constantly agitates against scientists who disagree with her.”).

To ensure the reliability of our coding, reply tweets were independently coded by two coders. Krippendorff’s alpha intercoder reliability was calculated using a random sample comprising 10% (n = 600) of the tweets, and the resulting values ranged between .70 and 1 (see supplementary material). Identifying forms of incivility, as well as trustworthiness concerns, within textual communication poses a significant challenge. Among these challenges, identifying sarcasm proves to be particularly difficult, especially within short forms of communication such as tweets with a limited character count. Nonetheless, a Krippendorff’s alpha coefficient of 0.70 for the sarcasm variable was considered acceptable.

5 Results

Overall, 6,000 reply tweets toward six German virologists were analyzed. Within this sample, 17.3% (n = 1039) of the replies contained at least one form of uncivil communication, while the other 82.7% of replies were considered civil as they did not contain a form of uncivil communication as per the provided definition of incivility (RQ₁). The majority of Twitter/X replies toward visible German virologists during the Covid-19 pandemic did not contain uncivil communication. This result aligns with previous research findings [e.g., Coe et al., 2014; Theocharis et al., 2020]. Roughly, in the same number of tweets (n = 1048), discussants raised trustworthiness concerns.

In order to acknowledge a continuum of incivility with milder and heavier forms, the occurrence of each incivility form individually was analyzed (see Table 2). Sarcastic or cynical language occurred by far most often (n = 416). Name-calling (n = 291) and emotionalized language (n = 165) were also identified often. Threats, silencing, sexism, or assigning stereotypes, as well as vulgarity, were less prevalent. Also, tweets that were considered uncivil, sometimes contained more than one form of incivility (M = 1.2, SD = 0.4). Thus, while the majority of tweets did not contain forms of incivility, a substantial number of tweets did contain one or more forms of uncivil communication (17.3%), with sarcasm being the most frequent and vulgarity the least frequent form identified within the 6,000 analyzed tweets.

The second research question encompassed whether the relative frequency of tweets containing uncivil communication and trustworthiness concerns increased across four periods. Table 1 presents the distribution of tweets with uncivil communication and trustworthiness concerns. The data indicate a partial, albeit inconsistent, increase.

The descriptive statistics show that the relative frequency of uncivil tweets in t₁ (16.03.2020–28.03.2020) is less than that of uncivil tweets in t₄ (03.03.2021–16.03.2021). However, the frequency does not increase consistently due to fewer uncivil tweets in t₃ (16.12.2020–03.01.2021) than in t₂ (28.10.2020–10.11.2020). Possible explanations for the lower shares in t₃ will be addressed in the discussion section.

The same descriptive analysis was conducted for the relative frequencies of tweets that contained concerns about the epistemic trustworthiness of scientists. Overall, concerns about the trustworthiness of scientists were raised in 17.5% of tweets across the four time periods. The development of relative frequencies within the four time periods is similar to the occurrence of incivility (see Table 1). The shares were lowest in t₁ and highest in t₄, while the tweets questioning trustworthiness were higher in t₂ than in t₃. Looking at the four time periods within the pandemic’s first year, it can be concluded that the occurrence of incivility and trustworthiness concerns increased over time, although not consistently.

Examining the individual forms of incivility over the four time periods individually, the inconsistent increase is also visible amongst most forms (Table 2). Sarcasm, name-calling and emotionalization are the three most frequent forms across all four time periods, with an exception in t₄ where silencing is occurring more often than emotionalization.

Moreover, this study investigated whether or not a relationship occurred between the prevalence of incivility and trustworthiness concerns within the replies toward German virologists during the pandemic (RQ₃). Table 3 shows the shares of incivility and trustworthiness concerns separately as well as together. The data show that 72.7% of tweets do not contain either incivility nor trustworthiness concerns. Solely incivility is prevalent in 9.8% of tweets, while solely trustworthiness concerns occur in 10.0% of the 6,000 tweets. In 7.5% of the tweets, incivility and trustworthiness concerns occur together.

A chi-square test of independence was performed to examine the relation between incivility and trustworthiness concerns. The relation between these variables was significant $\chi$² (1, N = 6,000) = 577.888, p < .001. Thus, there is a relationship between the co-occurrence of incivility and trustworthiness concerns within the reply tweets. However, it is important to note that this data does not establish a causal relationship in any direction.

6 Discussion

This study aimed to provide insight into the prevalence of uncivil communication and communicated concerns about the epistemic trustworthiness of scientists on Twitter/X. We analyzed reply tweets (N = 6,000) toward six visible German virologists during four different time periods within the first year of the Covid-19 pandemic.

These findings indicate that — independently considered — uncivil communication, as well as concerns about the trustworthiness of scientists, occur in the studied sample in more than every sixth Twitter/X reply to virologists’ tweets. The share of the occurrence of incivility is similar to what studies in news website comments have found [Coe et al., 2014] and similar to shares of incivility in tweets mentioning politicians [e.g., Theocharis et al., 2020]. This parallelism is noteworthy given the nature of the responsibilities for political decisions and the heightened legitimacy pressures that politicians contend with, in contrast to scientists. Nevertheless, the noteworthy similarity may be attributed to the perceived influence of scientists during the pandemic in shaping political decisions. Visible scientists frequently issued calls for political action [Biermann et al., 2023], and some were portrayed by the media as crisis managers [Jarren, 2020]. This may have amplified public’s perception that they exert influence on political decisions.

The frequency of various uncivil communication forms exhibits significant variations, with sarcastic or cynical language being the most prevalent by a substantial margin. At the same time, vulgarity is the least frequently employed form of incivility. The reasons behind the prevalence of the different forms can only be made speculatively. In general, social media communication is often more informalized, and especially on Twitter/X, restrictions on tweet length have been in place for a long time. Therefore, communication may be shorter and punchier [Ward & McLoughlin, 2020], possibly explaining high shares of emotionalization and sarcasm. Sarcasm incorporates a subtle attack within the discourse and often consists of provocative remarks that use exaggerated language or irony [Anderson & Huntington, 2017]. Here, sarcasm is considered as one form of incivility. Some researchers also view sarcasm and incivility as distinct concepts because subtle sarcasm might provoke incivility rather than being one of its manifestations [e.g., Anderson & Huntington, 2017]. In this study, sarcasm is considered one form of uncivil communication as it also incorporates an attack on deliberate discourse due to its provocative and derogatory language, which hinders the civil exchange of arguments. However, the high prevalence of sarcasm identified in this analysis shows the necessity to look at this form of uncivil language and its effects in more depth. Specifically, in the context of science communication, research on the dynamics of sarcasm is still considered limited [Anderson & Huntington, 2017].

Name-calling was the second most prevalent form of uncivil communication within the investigated tweets and is a predominant form of incivility that has also been the result of other studies. In their research analyzing news website comment sections, Coe et al. [2014] found name-calling to occur in 14% of the comments. In the present study, the overall prevalence of name-calling is 4.9% and thus significantly less than what Coe et al. [2014] identified. The name-calling often occurred within the investigated Twitter/X replies and might be explained partly by the nature of the material. Compared to news website comments, these comments were direct replies to scientists who may have presented study results but who also might have expressed their personal opinions on certain political measures. With the growing personalization tendencies that social media platforms and mass media enhance [Ward & McLoughlin, 2020], personal insults might be provoked more often.

The development of occurrences of incivility and concerns about trustworthiness should also be discussed. This study explored whether the share of incivility and the share of concerns about trustworthiness increased consistently over the four time periods. This was derived from an increasing polarization about certain political measures and governmental responses, as well as a slight decrease of trust in science as the pandemic endured, which was after a strong increase of trust in science at the beginning of the pandemic [Bromme et al., 2022]. Moreover, perceived scientific controversies about, for example, vaccine safety and prevention measures may have contributed to this. Surprisingly, the Twitter/X replies within the third time period contain less incivility and less trustworthiness concerns than in the second time period. The fourth time period again contains the highest share of incivility and trustworthiness concerns. Reasons for this might lie in the time frame of t₃ (16.12.2020–03.01.2021). An explanation that should be considered is that some polarizing debates paused or continued more civilized over the Christmas and New Year holidays. Moreover, during the holidays, the vaccine rollout started in Germany, which might have positively affected the sentiment of the discourse. That the same development can be found for incivility and trustworthiness concerns underscores this reasoning.

In addition, this study investigated a potential relationship between the occurrence of uncivil communication and concerns about epistemic trustworthiness. Results show a significant relationship between the occurrence of these two phenomena. This result partly underscores the findings of Anderson and Huntington [2017], who found that skeptical perspectives toward science tend to elicit a higher frequency of uncivil and sarcastic language. The results of this study also indicate that epistemic trustworthiness concerns are often communicated in an uncivil manner. Thus, integrating the occurrence of concerns about epistemic trustworthiness into investigations on incivility enhances the theoretical conceptualization by recognizing the varied manifestations of incivility. Therefore, future research should investigate this relationship within science communication on social media.

A few limitations must also be considered when drawing implications from this study. For this study, specific sampling requirements led to analyzing Twitter/X replies directed at just six selected scientists. The scientists needed to be full professors of virology and active researchers with a university affiliation as well as having over 10,000 followers. The number of followers was important to ensure these individuals communicated to a broader public instead of only within the scientific community. In addition, only reply tweets were analyzed; the contents of the original tweets were not considered. However, the preceding communication of the virologists may have influenced the replies as it may have fostered more emotional reactions depending on the statement type or language used. It also has to be acknowledged that the comments are not representative of recipients of the original tweets. It is however, important to investigate the prevalence of uncivil communication within tweets due to the effects that observing incivility can have. Moreover, the high epistemic uncertainty during the pandemic makes it almost impossible to assess the nature of the trustworthiness concern as new evidence questions the validity of earlier evidence or drawn implications that appeared on a regular basis. Thus, some trustworthiness concerns may have been justified or correct and may have also promoted the overall trust in science to enhance “informed trust” — also described as critical trust [Bromme, 2022].

7 Conclusion and outlook

This study highlights the importance of analyses concerning the responses toward scientists’ social media communication due to the potential of directly interacting with the public online which is accompanied by challenges and disruptions. This is especially true around a socio-scientific issue like the Covid-19 pandemic, which sparked controversial and polarized debates on topics such as mask-wearing and mandatory vaccination. By analyzing 6,000 reply tweets toward scientists from different time periods within the pandemic’s first year, this study provides insights into the dynamics and manifestations of online incivility in the context of science communication. As more than every sixth tweet contained uncivil language (17.3%), and more than every sixth tweet contained epistemic trustworthiness concerns (17.5%), this investigation shows high relevance. This study also provides insights into the frequencies of the different forms of uncivil communication, which is a valuable contribution as conceptualizations and operationalizations of incivility are divergent. Studies, investigating the effects of uncivil communication should test forms of incivility individually, as it is expected that effects and consequences for deliberation vary for different forms of incivility. This study shows that not only do the occurrence of forms of incivility differ strongly, but it is also visible that the occurrence inconsistently increases over the pandemic time period, which indicates growing controversiality in online debates around the pandemic on Twitter/X. In addition, a key finding is that epistemic trustworthiness concerns are intertwined with incivility, indicating that trustworthiness concerns are often communicated in an uncivil manner. These results enhance our understanding of the dynamics of online incivility during the pandemic insofar as incivility possibly has been affected by specific events (e.g., Christmas) or phases (e.g., increasing polarization). Moreover, as the development of occurrences of incivility and trustworthiness concerns resemble one another and co-occurrences exist, this study provides a first step into a nuanced understanding of how public sentiments toward visible scientists evolve during a global crisis.

Future research should investigate the effects of trustworthiness concerns that are communicated uncivilly and the differentiation when trustworthiness concerns are communicated civilly and when incivility occurs without addressing epistemic trustworthiness. Both phenomena may reinforce each other; however, the data do not allow us to make statements about whether uncivil comments affect trust attitudes or vice versa. Neither, we can make assumptions about whether scientists’ disagreements or statement types influenced the reply tweets. Former studies have shown that civil and uncivil disagreements amongst scientists can affect trust in science and scientific methods [Chinn & Hart, 2021]. Moreover, as the conceptualization of incivility is contested, in particular outside a purely political context, studies investigating this concept in more depth are needed. Therefore, content analyses and experimental studies should deal transparently with the underlying forms of their conceptualization of incivility and investigate potential differences of effects within the umbrella concept of incivility. Finally, research should further explore how both uncivil communication and raised trustworthiness concerns directed at scientists affect communication behavior, the overall motivation of scientists to engage publicly, and how scientists deal with these disruptions and challenges. This paper has offered a foundation for further research by providing evidence of the occurrences and manifestations of uncivil communication in the context of negotiating socio-scientific issues.

Acknowledgments

This research is part of the project "Science communication during pandemics: The role of public engagement in social media discussions", funded by Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - 458609429. Grant applicants are Monika Taddicken, Nicole Krämer and Stefan Stieglitz. I would like to thank Maximilian-Joseph Dichtl for his assistance with data coding as well as my colleagues and my supervisor Monika Taddicken at the Institute for Communication Science for their valuable feedback. I also appreciate the constructive comments from the anonymous reviewers.

References

Anderson, A. A., Brossard, D., Scheufele, D. A., Xenos, M. A. & Ladwig, P. (2013). The “nasty effect:” online incivility and risk perceptions of emerging technologies. Journal of Computer-Mediated Communication 19(3), 373–387. doi:10.1111/jcc4.12009

Anderson, A. A. & Huntington, H. E. (2017). Social media, science, and attack discourse: how Twitter discussions of climate change use sarcasm and incivility. Science Communication 39(5), 598–620. doi:10.1177/1075547017735113

Barnes, R. M., Johnston, H. M., MacKenzie, N., Tobin, S. J. & Taglang, C. M. (2018). The effect of ad hominem attacks on the evaluation of claims promoted by scientists. PLOS ONE 13(1), e0192025. doi:10.1371/journal.pone.0192025

Biermann, K., Peters, N. & Taddicken, M. (2023). “You can do better than that!”: Tweeting scientists addressing politics on climate change and Covid-19. Media and Communication 11(1). doi:10.17645/mac.v11i1.5961

Borah, P. (2012). Does it matter where you read the news story? Interaction of incivility and news frames in the political blogosphere. Communication Research 41(6), 809–827. doi:10.1177/0093650212449353

Bormann, M., Tranow, U., Vowe, G. & Ziegele, M. (2021). Incivility as a violation of communication norms—a typology based on normative expectations toward political communication. Communication Theory 32(3), 332–362. doi:10.1093/ct/qtab018

Broer, I. & Hasebrink, U. (2022). Wissenschaftskommunikation als kommunikative Figuration. Ein konzeptioneller Rahmen für die empirische Untersuchung von Domänen der Wissenschaftskommunikation. [Science communication as communicative figuration. A conceptual framework for the empirical investigation of science communication’s multiple domains]. Medien & Kommunikationswissenschaft 70(3), 234–255. doi:10.5771/1615-634x-2022-3-234

Bromme, R. (2022). Informiertes Vertrauen in Wissenschaft: Lehren aus der Covid-19 Pandemie für das Verständnis naturwissenschaftlicher Grundbildung (scientific literacy) [Informed trust in science: lessons from the Covid-19 pandemic for the conceptualization of science literacy]. Unterrichtswissenschaft 50(3), 331–345. doi:10.1007/s42010-022-00159-6

Bromme, R., Mede, N. G., Thomm, E., Kremer, B. & Ziegler, R. (2022). An anchor in troubled times: trust in science before and within the Covid-19 pandemic. PLOS ONE 17(2), e0262823. doi:10.1371/journal.pone.0262823

Brooks, D. J. & Geer, J. G. (2007). Beyond negativity: the effects of incivility on the electorate. American Journal of Political Science 51(1), 1–16. doi:10.1111/j.1540-5907.2007.00233.x

Chinn, S. & Hart, P. S. (2021). Can’t you all just get along? Effects of scientific disagreement and incivility on attention to and trust in science. Science Communication 44(1), 108–129. doi:10.1177/10755470211054446

Cinelli, M., Quattrociocchi, W., Galeazzi, A., Valensise, C. M., Brugnoli, E., Schmidt, A. L., … Scala, A. (2020). The Covid-19 social media infodemic. Scientific Reports 10(1). doi:10.1038/s41598-020-73510-5

Coe, K., Kenski, K. & Rains, S. A. (2014). Online and uncivil? Patterns and determinants of incivility in newspaper website comments. Journal of Communication 64(4), 658–679. doi:10.1111/jcom.12104

Collins, L. & Nerlich, B. (2014). Examining user comments for deliberative democracy: a corpus-driven analysis of the climate change debate online. Environmental Communication 9(2), 189–207. doi:10.1080/17524032.2014.981560

Dieckmann, N. F., Johnson, B. B., Gregory, R., Mayorga, M., Han, P. K. J. & Slovic, P. (2015). Public perceptions of expert disagreement: bias and incompetence or a complex and random world? Public Understanding of Science 26(3), 325–338. doi:10.1177/0963662515603271

Dohle, S., Wingen, T. & Schreiber, M. (2020). Acceptance and adoption of protective measures during the Covid-19 pandemic: the role of trust in politics and trust in science. Social Psychological Bulletin 15(4). doi:10.32872/spb.4315

Drescher, L. S., Roosen, J., Aue, K., Dressel, K., Schär, W. & Götz, A. (2021). The spread of Covid-19 crisis communication by german public authorities and experts on Twitter: quantitative content analysis. JMIR Public Health and Surveillance 7(12), e31834. doi:10.2196/31834

Egelhofer, J. L. (2023). How politicians’ attacks on science communication influence public perceptions of journalists and scientists. Media and Communication 11(1), 361–373. doi:10.17645/mac.v11i1.6098

Gervais, B. T. (2015). Incivility online: affective and behavioral reactions to uncivil political posts in a web-based experiment. Journal of Information Technology & Politics 12(2), 167–185. doi:10.1080/19331681.2014.997416

Giddens, A. (2007). The consequences of modernity. (Original work published 1990). Stanford University Press.

Gierth, L. & Bromme, R. (2019). Attacking science on social media: how user comments affect perceived trustworthiness and credibility. Public Understanding of Science 29(2), 230–247. doi:10.1177/0963662519889275

Goodell, R. (1977). The visible scientists. The Sciences 17(1), 6–9. doi:10.1002/j.2326-1951.1977.tb01494.x

Habermas, J. (1985). The theory of communicative action. Beacon press.

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:10.1371/journal.pone.0139309

Jarren, O. (2020). Im Krisenmodus: Das öffentlich-rechtliche Fernsehen in Zeiten von Corona [In crisis mode: public television in times of Corona]. epd Medien 13, 3–6. Retrieved from https://www.zora.uzh.ch/id/eprint/186723/1/jarren_corona.pdf

Jiang, X., Su, M.-H., Hwang, J., Lian, R., Brauer, M., Kim, S. & Shah, D. (2021). Polarization over vaccination: ideological differences in Twitter expression about Covid-19 vaccine favorability and specific hesitancy concerns. Social Media + Society 7(3), 205630512110484. doi:10.1177/20563051211048413

Kata, A. (2012). Anti-vaccine activists, Web 2.0, and the postmodern paradigm—an overview of tactics and tropes used online by the anti-vaccination movement. Vaccine 30(25), 3778–3789. doi:10.1016/j.vaccine.2011.11.112

König, L. & Jucks, R. (2019). Hot topics in science communication: aggressive language decreases trustworthiness and credibility in scientific debates. Public Understanding of Science 28(4), 401–416. doi:10.1177/0963662519833903

Lang, J., Erickson, W. W. & Jing-Schmidt, Z. (2021). #MaskOn! #MaskOff! Digital polarization of mask-wearing in the United States during Covid-19. PLOS ONE 16(4), e0250817. doi:10.1371/journal.pone.0250817

Laslo, E. & Baram-Tsabari, A. (2019). Expressions of ethics in reader comments to animal experimentation and climate change online coverage. International Journal of Science Education, Part B 9(4), 269–284. doi:10.1080/21548455.2019.1654145

Lu, H., Chu, H. & Ma, Y. (2021). Experience, experts, statistics, or just science? Predictors and consequences of reliance on different evidence types during the Covid-19 infodemic. Public Understanding of Science 30(5), 515–534. doi:10.1177/09636625211009685

Luhmann, N. (2014). Vertrauen: Ein Mechanismus der Reduktion sozialer Komplexität [Trust: A mechanism for reducing social complexity]. UTB GmbH.

Martini, C., Battisti, D., Bina, F. & Consolandi, M. (2022). Knowledge brokers in crisis: public communication of science during the Covid-19 pandemic. Social Epistemology 36(5), 656–669. doi:10.1080/02691728.2022.2116961

Marwick, A. E. & boyd, d. (2010). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society 13(1), 114–133. doi:10.1177/1461444810365313

Masullo Chen, G. & Lu, S. (2017). Online political discourse: exploring differences in effects of civil and uncivil disagreement in news website comments. Journal of Broadcasting & Electronic Media 61(1), 108–125. doi:10.1080/08838151.2016.1273922

Masullo Chen, G., Muddiman, A., Wilner, T., Pariser, E. & Stroud, N. J. (2019). We should not get rid of incivility online. Social Media + Society 5(3), 205630511986264. doi:10.1177/2056305119862641

Mayer, R. C., Davis, J. H. & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review 20(3), 709–734. doi:10.5465/amr.1995.9508080335

Muddiman, A. (2017). Personal and public levels of political incivility. International Journal of Communication 11, 3182–3202. Retrieved from https://ijoc.org/index.php/ijoc/article/view/6137

Neuberger, C., Bartsch, A., Fröhlich, R., Hanitzsch, T., Reinemann, C. & Schindler, J. (2023). The digital transformation of knowledge order: a model for the analysis of the epistemic crisis. Annals of the International Communication Association 47(2), 180–201. doi:10.1080/23808985.2023.2169950

Nogrady, B. (2021). ‘I hope you die’: how the Covid pandemic unleashed attacks on scientists. Nature 598(7880), 250–253. doi:10.1038/d41586-021-02741-x

Nölleke, D., Leonhardt, B. M. & Hanusch, F. (2023). “The chilling effect”: medical scientists’ responses to audience feedback on their media appearances during the Covid-19 pandemic. Public Understanding of Science 32(5), 546–560. doi:10.1177/09636625221146749

Olesk, A. (2021). The types of visible scientists. JCOM 20(02), A06. doi:10.22323/2.20020206

Papacharissi, Z. (2004). Democracy online: civility, politeness, and the democratic potential of online political discussion groups. New Media & Society 6(2), 259–283. doi:10.1177/1461444804041444

Pascual-Ferrá, P., Alperstein, N., Barnett, D. J. & Rimal, R. N. (2021). Toxicity and verbal aggression on social media: polarized discourse on wearing face masks during the Covid-19 pandemic. Big Data & Society 8(1), 205395172110235. doi:10.1177/20539517211023533

Popper, K. R. (2003). Science: Conjectures and refutations. In A. Bird & J. Ladyman (Eds.), Arguing About Science (pp. 15–43). Routledge.

Rauchfleisch, A., Siegen, D. & Vogler, D. (2021). How Covid-19 displaced climate change: mediated climate change activism and issue attention in the Swiss media and online sphere. Environmental Communication 17(3), 313–321. doi:10.1080/17524032.2021.1990978

Reif, A. (2021). Mehr Raum für Vertrauen? Potenzielle Veränderungen des Vertrauens in Wissenschaft durch partizipative Onlineumgebungen [More space for trust? Potential changes of trust in science through participatory online environments]. In T. Döbler, C. Pentzold & C. Katzenbach (Eds.), Räume digitaler Kommunikation [Spaces of digital communication] (pp. 210–243). Herbert von Halem.

Reif, A. & Guenther, L. (2021). 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. doi:10.1080/21515581.2022.2075373

Rösner, L., Winter, S. & Krämer, N. C. (2016). Dangerous minds? Effects of uncivil online comments on aggressive cognitions, emotions, and behavior. Computers in Human Behavior 58, 461–470. doi:10.1016/j.chb.2016.01.022

Sadler, T. D., Barab, S. A. & Scott, B. (2007). What do students gain by engaging in socioscientific inquiry? Research in Science Education 37(4), 371–391. doi:10.1007/s11165-006-9030-9

Southern, R. & Harmer, E. (2019). Twitter, incivility and “everyday” gendered othering: an analysis of tweets sent to UK members of parliament. Social Science Computer Review 39(2), 259–275. doi:10.1177/0894439319865519

Stieglitz, S., Mirbabaie, M., Ross, B. & Neuberger, C. (2018). Social media analytics—challenges in topic discovery, data collection, and data preparation. International Journal of Information Management 39, 156–168. doi:10.1016/j.ijinfomgt.2017.12.002

Su, L. Y.-F., Xenos, M. A., Rose, K. M., Wirz, C., Scheufele, D. A. & Brossard, D. (2018). Uncivil and personal? Comparing patterns of incivility in comments on the Facebook pages of news outlets. New Media & Society 20(10), 3678–3699. doi:10.1177/1461444818757205

Sydnor, E. (2017). Platforms for incivility: examining perceptions across different media formats. Political Communication 35(1), 97–116. doi:10.1080/10584609.2017.1355857

Taddicken, M. & Krämer, N. (2021). Public online engagement with science information: on the road to a theoretical framework and a future research agenda. JCOM 20(03), A05. doi:10.22323/2.20030205

Theocharis, Y., Barberá, P., Fazekas, Z. & Popa, S. A. (2020). The dynamics of political incivility on Twitter. SAGE Open 10(2), 215824402091944. doi:10.1177/2158244020919447

Theocharis, Y., Barberá, P., Fazekas, Z., Popa, S. A. & Parnet, O. (2016). A bad workman blames his tweets: the consequences of citizens’ uncivil Twitter use when interacting with party candidates. Journal of Communication 66(6), 1007–1031. doi:10.1111/jcom.12259

Utz, S., Gaiser, F. & Wolfers, L. N. (2022). Guidance in the chaos: effects of science communication by virologists during the Covid-19 crisis in Germany and the role of parasocial phenomena. Public Understanding of Science 31(6), 799–817. doi:10.1177/09636625221093194

Vosoughi, S., Roy, D. & Aral, S. (2018). The spread of true and false news online. Science 359(6380), 1146–1151. doi:10.1126/science.aap9559

Ward, S. & McLoughlin, L. (2020). Turds, traitors and tossers: the abuse of UK MPs via Twitter. The Journal of Legislative Studies 26(1), 47–73. doi:10.1080/13572334.2020.1730502

Wintterlin, F., Hendriks, F., Mede, N. G., Bromme, R., Metag, J. & Schäfer, M. S. (2022). Predicting public trust in science: the role of basic orientations toward science, perceived trustworthiness of scientists, and experiences with science. Frontiers in Communication 6. doi:10.3389/fcomm.2021.822757

Wissenschaft im Dialog (2023, October 10). Wissenschaftsbarometer. Eine repräsentative Bevölkerungsumfrage zu Wissenschaft und Forschung [Science barometer: a representative survey on science and research]. Retrieved from https://www.wissenschaft-im-dialog.de/projekte/wissenschaftsbarometer/

Notes

^1. As of July 24th, 2023, Elon Musk, owner of Twitter Inc. (now X Corp.), renamed the platform to "X." Thus, in this paper, the platform will be referred to as Twitter/X. Along with the ownership change came changes to the functionalities of the platform. This is to be considered when drawing implications from the results of this study.

About the author

Nicola Peters is a research assistant at the Institute for Communication Science at the Technische Universität Braunschweig, Germany. She holds a MA degree in digital media and society from the University of Bremen, Germany. Currently, she is working on her PhD project on incivility in online discussions around socio-scientific issues.

E-mail: nicola.peters@tu-braunschweig.de

Supplementary material

Available at https://doi.org/10.22323/2.23060203
Codebook for the content analysis.