Volume 4 Issue 2, Article
Tome 4 numéro 2, Article

Sibo Chen, Nicole Blanchett, and S. Lecourt

Abstract

As anti-press hostility intensifies globally, its impact on academics, whose presence as authors in journalism-based publications has increased in recent years, presents an understudied subject in communication and media studies literature. Approaching this subject using data from a survey of 732 authors affiliated with The Conversation Canada, a news startup specializing in explanatory journalism, this mixed-methods case study examined how academic authors perceive and cope with toxic comments and varying forms of anti-press hostility. Survey data are contextualized with qualitative interviews. The findings suggest that toxic comments pose a notable challenge to practice at The Conversation Canada. These comments are leading to self- censorship by causing a number of the affected authors to avoid expressing their viewpoints or writing about certain topics, as well as to reconsider their knowledge mobilization efforts. In sum, this study contributes to the body of literature on anti-press hostility and informs relevant public and policy discussions on the need to mitigate polarization online.

Keywords: anti-press hostility, dark participation, digital trolling, harassment, hostility towards the press, online hate, toxic comments, The Conversation Canada, journalism

Résumé

Alors que l’hostilité envers la presse s’intensifie à l’échelle mondiale, son impact sur les universitaires, dont la présence en tant que auteur.es dans des publications basées sur le journalisme a augmenté ces dernières années, constitue un sujet sous-étudié dans la littérature sur la communication et les études médiatiques. Abordant ce sujet à l’aide de données provenant d’une enquête menée auprès de 732 auteurs affiliés à The Conversation Canada, une entreprise de presse spécialisée dans le journalisme explicatif, cette étude de cas à méthodes mixtes a examiné la manière dont les auteur.es universitaires perçoivent les commentaires toxiques et les diverses formes d’hostilité envers la presse. Les données de l’enquête sont mises en contexte par des entretiens qualitatifs. Les résultats suggèrent que les commentaires toxiques posent un défi important à la pratique de The Conversation Canada. Ces commentaires conduisent à l’autocensure en amenant un certain nombre d’auteur.es concernés à éviter d’exprimer leur point de vue ou d’écrire sur certains sujets, ainsi qu’à reconsidérer leurs efforts de mobilisation des connaissances. En résumé, cette étude contribue à l’ensemble de la littérature sur l’hostilité anti-presse et alimente les discussions publiques et politiques pertinentes sur la nécessité d’atténuer la polarisation en ligne.

Mots-clés : hostilité anti-presse, participation obscure, trolling numérique, harcèlement, hostilité envers la presse, haine en ligne, commentaires toxiques, The Conversation Canada, journalisme

ARTICLE

---

The dark side of public visibility: How academic
authors perceive and cope with anti-press hostility

Sibo Chen, Nicole Blanchett, and S. Lecourt

 

INTRODUCTION

In communication and media studies, anti-press hostility, also known as “hostility towards the press” (Miller, 2023a, p. 1230), is a well-established research topic whose specifics have been characterized by a variety of terms, including abuse, aggression, bullying, hostility, incivility, violence, and mobbing (e.g., Cheruiyot, 2022; Kim & Shin, 2022; Miller, 2023a; Waisbord, 2020). As these terms suggest, anti-press hostility encompasses a broad spectrum of attitudes, expressions, or actions that are antagonistic towards journalists and media organizations. In this study, we specifically focus on online harassment in the form of toxic remarks found in the comment sections of news websites, and those sent directly to authors through social media and emails.

The deteriorating online environment affects not only professional journalists but also others engaged in journalistic work. For example, a rising demand for knowledge mobilization in academia, resulting from collective pushes by funding agencies, universities, and the public, prompts many academics to write opinion and analysis pieces or other forms of journalistic content. The increased public visibility brought about by these articles, however, exposes their authors to the same toxic comments and other forms of online harassment faced by journalists (Glassey, 2017), an inevitable consequence of increased engagement with larger audiences (Gosse et al., 2021; Oksanen et al., 2022).

Academics engaging in public scholarship are often subjected to the same kinds of vitriol faced by journalists even when they aren’t engaging in “journalism” per se. In particular, the COVID-19 pandemic and its associated “infodemic” have further polarized public discourse and increased hostility towards those sharing their expertise and research-based insights (“COVID scientists in the public eye,” 2021; Wright et al., 2022). The drivers of the harassment and intimidation of academics are complex, widespread, and global, as indicated by a growing body of research and news stories (e.g., Frangou, 2019; Gewin, 2018; Gosse et al., 2021; Nogrady, 2024; Oksanen et al., 2022). This trend underscores the pressing necessity to comprehend the intimidation of public communicators, which is a burgeoning issue that negatively impacts democratic discourse.

Following Miller and Lewis (2022), we define harassment as “unwanted behaviors [in both verbal and physical forms] that are sexual, abusive, sexist or aggressive in nature” (p.81). There is a concerning trend of escalating hostility toward the press worldwide. In North America, for instance, the pervasiveness and grave consequences of this issue have been widely documented (e.g., Canada Press Freedom Project, 2024; Eschner, 2022; Holton et al., 2023; Lewis et al., 2020; Miller, 2021; Miller & Kocan, 2022; Waisbord, 2020). Journalists are vulnerable to various forms of harassment (sexist, racist, violent, etc.) that have a wide range of negative effects. Similar trends are also observed in other parts of the world, irrespective of their levels of press freedom (Chen et al., 2020; Kim & Shin, 2022; Löfgren Nilsson & Örnebring, 2016; Obermaier et al., 2018; Posetti et al., 2021; Tandoc et al., 2021). In short, anti-press hostility presents a sprawling and multifaceted issue with distinct nuances at both the local and national levels (Miller, 2023; Waisbord, 2022a). Its alarming prevalence not only endangers the autonomy and safety of journalists but also imperils the anticipated role of journalism in a democratic society.

The  issue  also  raises  important  questions concerning journalist-audience relationships in the digital era. When digital journalism initially gained popularity in the early 2000s, the seemingly endless possibilities it offered for interactions between journalists and audiences led to optimism among scholars and practitioners. There was belief that greater citizen participation in the news process would rejuvenate journalism and strengthen democracy (Lewis et al., 2020; Quandt, 2018; Westlund, 2020).

In tandem with the rapid expansion of digital media, however, the optimistic outlook was dimmed by “dark participation,” namely “negative, selfish, or even deeply sinister forms of audience participation, such as ‘trolling,’ (Coles & West, 2016; Mihaylov & Nakov, 2016) strategic ‘piggybacking’ on journalistic reputation, and large-scale disinformation in uncontrolled news environments” (Quandt, 2018, p. 40). As online news consumption continues to grow, journalists’ performances are increasingly measured by their ability to interact with audiences on social media platforms (Davis Kempton & Connolly-Ahern, 2022; Blanchett, 2021; Tandoc et al., 2021). Yet, these interactions are frequently marred by dark participation, impacting editorial choices and exacerbating the difficulties faced by journalists (Blanchett et al., 2024).

Given the increasing presence of academics as news authors, the scope and nature of anti-press hostility they encounter presents an understudied subject in communication and media studies literature. This article aims to address this gap by conducting an exploratory study focusing on authors affiliated with The Conversation Canada, a news startup specializing in explanatory journalism and exclusively publishing articles authored by academics. Using a mixed-methods approach that combines surveys and interviews, the study sheds light on how academics perceive and cope with toxic comments and other forms of anti-press hostility, thereby contributing to policy discussions on the need to mitigate the escalation of factors such as societal polarization across Western democracies (World Economic Forum, 2024).

LITERATURE REVIEW

The literature review begins with an overview of research on anti-press hostility and then explores its connection to academic work and academics writing in journalistic spaces. The launch of the UN Plan of Action on the Safety of Journalists and the Issue of Impunity in 2015 was a major catalyst for renewed global scholarly and public interest in anti-press hostility (Fadnes et al., 2019; Miller, 2023). While significant attention has been paid to how journalists’ safety remains a long-standing problem in the Global South, underscored by concerning cases of arbitrary detention, forced disappearances, kidnapping, torture, and murder (Reporters without Borders, 2023; Waisbord, 2022a), there is an expanding body of research exploring what exacerbates journalists’ experiences with various forms of hostility in democratic and semi-democratic countries (e.g., Chen et al., 2020; Eschner, 2022; Löfgren & Örnebring, 2016; Miller, 2021; Posetti et al., 2021; Stahel & Schoen, 2020; Tandoc et al., 2021).

Anti-press hostility: Structural factors

A complex web of factors contributes to escalating hostility toward the press in recent years. For example, the organizational contexts in which journalists work can either aggravate or ameliorate the anxiety, fatigue, and frustration they experience when targeted by harassment (Westlund et al., 2022). Unfortunately, research also shows that many news organizations commonly fail to make systematic efforts to thwart hostile interactions with external actors (Holton et al., 2023). Some scholars attribute this deficiency to media managers’ vague understanding of online harassment or underestimation of its severity, while others argue that the importance of online traffic to news organizations’ economic performance today discourages them from imposing stringent measures regulating user-generated content (Hiltunen, 2022; Malcorps et al., 2022). Regardless of the causes, journalism scholars agree that the absence of formalized protocols on such interactions at many news organizations leads to an individualistic, case-by-case approach to the handling of harassment incidents, failing to recognize such incidents as a challenge requiring comprehensive institutional support (Pearson et al., 2022; Nelson, 2022; Waisbord, 2022a; Westlund et al., 2022).

Internet culture, in addition to organizational contexts, is another major structural factor shaping the dynamics of anti-press hostility. The past decade has witnessed the amplification of online toxicity marked by aggressive behaviors, hate speech, and incivility. Quandt (2018) refers to these negative forms of digital citizen participation as “dark participation” (p. 36) and associates their prevalence with the recent populist wave in Western democracies. In accordance with this perspective, scholars such as Hiltunen (2022), Waisbord (2020), and Westlund (2021) suggest that the surge of dark participation is fueled by dis/misinformation, societal divides, and political polarization,  which  reflects  the  fragmentation of the public sphere evident today. Additionally, research shows that platform affordances enable emotionally-laden content, particularly that which expresses anger, to move faster and more widely in online environments (Paletz et al., 2023).

Dark participation strategically targets social media and news comment sections because their existing large audience base offers agitators and trolls with immediate visibility without any formidable barriers. Meanwhile, as social media enhances the digital publicity of journalists—which shifts public attention from journalism to those doing the reporting— journalists, and anyone doing journalism work, face growing threats from individuals who view news organizations as institutions propagating ideologies they strongly disagree with (Brin & Charlton, 2022). In short, the mounting challenge posed by dark participation to journalism not only prompts us to reassess our understanding of what constitutes legitimate media criticism, but also highlights how the concept of free speech is increasingly manipulated by members of the far-right, who position themselves as persistent and vocal critics of mainstream media—a strategy that fuels cynicism towards such outlets (Broockman & Kalla, 2022; Carlson et al., 2021; Figenschou & Ihlebæk, 2019).

Anti-press hostility and journalist identities

Journalists are not equally impacted by anti-press hostility; gender, race/ethnicity, religion, and disability are factors in who is targeted (Chen et al., 2020; Davis Kempton & Connolly-Ahern, 2022; Lewis & Coddington, 2020; Miller, 2023a; Waisbord, 2020). A prime illustration of this pattern is that women journalists are more likely to face intimidation and harassment compared to their male counterparts. As Miller and Lewis (2022) summarize, women journalists face four main types of harassment: “(1) disruptive in-person harassment, (2) physical and abrasive in-person harassment, (3) online harassment as unwanted sexual advances, and (4) online harassment as threats and criticisms” (p. 87). Such harassment, besides sexually objectifying women journalists, trivializes their professional accomplishments, resulting in significant repercussions on individual, interpersonal, and professional levels.

The research conducted by Barão da Silva et al. (2022), for instance, revealed how a thriving hostile environment affects daily workflow, disrupts family routines, and causes emotional distress among Brazilian women journalists. Many women journalists also report that safety and privacy concerns place them in a quandary: either they continue to engage with audiences, thereby exposing themselves to potential online abuse, or they choose self-protection by regularly blocking suspicious social media users (Miller, 2023a; Waisbord, 2022b). The latter inevitably reduces audience engagement, an outcome that could be detrimental to their careers.

Additionally, there are observed gender and cultural differences in how journalists cope with harassment. Miller (2023b) and Stahel and Schoen (2020), among others, have suggested that women journalists tend to employ avoidance strategies more frequently than male journalists. While, as noted by these scholars, there is a possible difference in behavior related to gender-role socialization, there is also the issue of the severity of toxicity in comments received. Online attacks directed at women are considerably more hateful and violent than those experienced by men (Posetti et al., 2021); therefore, it would be logical to avoid such exposure. Compounding the nuances of journalists’ coping strategies are cultural factors. In cultures where societal norms discourage public confrontation with masculinity, women journalists are hindered from leveraging personal social media channels to share their experiences (Barão da Silva et al., 2022; Chen et al., 2020; Tandoc et al., 2021).

The impacts of anti-press hostility

Journalists’ exposure to online and offline harassment has significant implications not only on their mental health but also on their work attitudes. Journalists report that harassment induces negative emotions such as discomfort, anger, lethargy, and fear, which in turn drives certain work-related behaviors such as avoiding interviews, reducing social media presence, and even considering leaving the profession (Lewis et al., 2020; Miller, 2021; Pearson et al., 2022).

In more severe cases where journalists have been assaulted while covering street protests, for example, the result can be severe anxiety and post-traumatic stress disorder, further reinforcing the desire to leave journalism (Miller & Kocan, 2022). Harassment  also influences the interaction dynamics among journalists themselves, as well as between journalists and their audiences. A study focusing on South Korean journalists (Kim & Shin, 2022) revealed that with minimal institutional support, journalists manage the overwhelming negative emotions caused by harassment using emotion-focused coping mechanisms, such as striving for perfectionism in their daily work, blaming fellow journalists, creating emotional distance from   audiences, and expressing resentment against readers. Similarly, Lewis and his colleagues (2020) argue that “the more often a journalist is harassed online, the more likely they are to take a dim view of their audiences across several key dimensions: that is, to see their audiences as less rational and unlike themselves, and to see interaction with their audiences as less valuable” (p. 1062).

Meanwhile, anti-press hostility fosters a climate of fear within news organizations, potentially undermining the autonomy and editorial independence of journalists. This contradicts a fundamental premise prevalent in previous research on reciprocal journalism, namely that increased interactions between journalists and the public could promote mutual trust, benefit sharing, and community formation (Deavours et al., 2022; Lewis et al., 2020; Quandt, 2018). In practice, however, this premise is substantially compromised by various forms of dark participation, prompting questions about what constitutes legitimate media criticism (Carlson et al., 2021; Cheruiyot, 2022; Devours et al., 2022; Figenschou & Ihlebæk, 2019).

A concerning trend that reflects these questions is the rise of “mob censorship” (Miller, 2021a; Waisbord, 2020, 2022a; Westlund et al., 2022), as exemplified by populist leaders’ persistent demonization of journalists and mainstream news media. Mob censorship, as a violent form of public speech intended to intimidate opponents into silence, underscores the precarious nature of upholding effective democratic speech rights in the face of looming polarization. The growing threat that mob censorship poses to journalism demonstrates the unsettling reality that censorship is no longer the exclusive domain of powerful entities such as governments and corporations seeking to impede freedom of expression. To mitigate mob censorship, it’s crucial to distinguish between criticism and cynicism. Ideally, criticism is a rational act aimed at improving current conditions, while cynicism is generally an emotionally-driven judgement meant to breed distrust (Waisbord, 2020). In practice, the majority of alleged “media criticism” from far-right groups merely serves as a pretext for cynical assaults intended to provoke antagonism towards mainstream news media (Cheruiyot, 2022; Figenschou & Ihlebæk, 2019).

In sum, safeguarding journalists, and those who do journalistic work, from online and offline harassment extends beyond mere employment rights issues—it constitutes matters of human rights and press freedom, which significantly impact the quality of information circulated in the public sphere and the diversity of people sharing that information (Cheruiyot, 2022; Kim & Shin, 2022; Miller, 2021a; Waisbord, 2022a).

Hostility targeting academics

The current literature on anti-press hostility primarily focuses on the experiences of professional journalists, without paying much attention to the increasing number of academics engaging in journalistic work. Considering this gap, this research explores the extent to which anti-press hostility curtails scholars’ enthusiasm for public writing. Online hostility targeting journalists and academics are related phenomena that reflect intensifying anti-establishment and anti-intellectual sentiments in the public sphere (Waisbord, 2020; Wright et al., 2022). Although academics, whether featured as experts in news stories or as authors of opinion or analysis pieces, tend to not identify themselves as “journalists,” they are often perceived by the public as “elite voices.” This perception aligns them with journalists in the eyes of many, resulting in similar experiences of harassment and hostility when they participate in public conversations (“COVID scientists in the public eye,” 2021).

Previous research on online harassment targeting academics identified causes and consequences that parallel those observed in studies of anti-press hostility. For instance, investigations by Gosse et al. (2019) and Oksanen et al. (2022) have shown that online harassment is deeply intertwined with the identity and work of scholars. Such harassment and violence can also seep into physical spaces in academia, as seen in a recent stabbing in a University of Waterloo classroom (MacDonald & Kalentsis, 2023). The pressures stemming from such harassment can lead to significant psychological distress, comparable to the experiences reported by journalists facing hostility (Lewis et al., 2020; Miller & Kocan, 2022; Pearson et al., 2022). These findings highlight a troubling overlap in the challenges that both groups face as they navigate their roles as public figures.

Moreover, academics, like journalists, frequently report a lack of institutional support when targeted by online hostility (Holton et al., 2023; Nogrady, 2024). This exacerbates their vulnerability to harassment, rendering them less equipped to cope with the fallouts from their knowledge mobilization activities. Thus, in today’s digital landscape, where online harassment is increasingly characterized by organized mob censorship rather than isolated acts of aggression, it is important to document and analyze the distinctive experiences and challenges encountered by academics who face increased opportunities of exposure to harassment when they venture into journalistic spaces.

DATA AND METHODS

This study focuses on the experiences of authors affiliated with The  Conversation Canada (https://theconversation.com/ca), a member of a global network of non-profit news organizations (The Conversation network) that specialize in publishing research-based articles in collaboration with scholars across a wide range of universities. In Canada, it is published by the Academic Journalism Society. The subsequent sections will offer a synopsis of The Conversation network’s journalism practice, followed by a detailed account of the data collection and analysis procedures.

The Conversation network

What distinguishes The Conversation network members apart from legacy news media are two features. First, they only publish articles authored by academics. A prospective author must be associated with a university. They also must demonstrate their research expertise relevant to the story they pitch or have been asked to write during preliminary discussions with a handling editor. The second feature lies in the depth and breadth of Conversation articles, many of which synthesize findings from multiple empirical studies to explain complex and high-profile issues. During the COVID-19 pandemic, for instance, the network published not only authoritative explanations of the scientific basis for public health measures written by epidemiologists, but also critical analyses of the pandemic’s sociocultural effects written by humanities scholars (Chen, 2022).

As evidenced by both features, The Conversation network is a practitioner of explanatory journalism. Similar to other explanatory news websites such as FiveThirtyEight and Vox, the network’s global expansion benefits from two trends that have recently boosted the popularity of explanatory news (Chen, 2022; Hermida & Young, 2019; Mann, 2016). First, there is a growing public interest in in-depth news pieces that elucidate the complicated concepts underpinning high-profile public issues. Second, the availability of data analysis and visualization applications (e.g., Google Spreadsheet and Tableau) significantly reduces the technical barriers to data-driven storytelling that makes complex topics and datasets easier to understand.

Departing from the practices of legacy news media where academics mainly hold the peripheral role as sources of quotations, The Conversation network brands itself as an innovator. It implements a “scholars as journalists” model that integrates academics into the news production process. In this model, “researchers assume the roles of journalists, undertaking traditional tasks such as pitching and writing a story” (Hermida & Young, 2019, p. 94). Concurrently, the primary responsibility of The Conversation network’s editorial teams is to aid scholars in creating content that has “journalistic flair.”

By promoting reciprocity between journalists and academics, The Conversation network seeks to transform universities into what could be described as “giant newsrooms.” Its establishment reflects the ongoing deterritorialization of media space and the evolving public perceptions of what constitutes journalism practice (Blanchett et al., 2025; Deuze & Witschge, 2018). This not only enables the publication of articles on topics requiring a great deal of specialized knowledge (e.g., epidemiology, economics, virology, etc.), but also harnesses academic expertise to strengthen the credibility of Conversation articles against dis/ misinformation (Allen & Lucky, 2023).

Moreover, all Conversation articles are published under a Creative Commons license, enabling news organizations worldwide to republish them. In Canada, for instance, top republishers of Conversation articles include the websites of well-known legacy media such as the National Post, Global News, and Maclean’s. This promotes the dissemination and outreach of research for the public good.

The above characteristics position The Conversation network as a promising force in global journalism innovation, and its efforts to combat dis/misinformation could be a factor in increased public recognition. From July 2020 to June 2021, when the COVID-19 pandemic ignited a significant public demand for expert opinions, The Conversation United States (2022) alone attracted 68 million reads per month (including republication in other media). As of the end of 2023, it had accumulated “over 1.1 billion pageviews since launch” (The Conversation United States, 2024, p. 4).

In 2023, The Conversation Canada and its Francophone edition “published about 1,400 news and analysis articles . . . written by 1,700 academic experts from Canadian universities; and these articles were viewed over 32 million times by readers” (Varano, 2024 January 1, para. 2). Yet, accompanying this substantial rise in digital publicity is a pressing concern: the pervasive anti-press hostility now affects Conversation authors in a variety of ways.

In late 2022, for instance, The Conversation Canada was inundated with toxic comments that severely impeded direct and reciprocal conversations between academics and readers. Trolls and agitators overwhelmed the website’s small editorial team, forcing it to implement the policy that “comments are only open on selected articles and are typically open for 72hrs” (The Conversation Canada, 2021, para 3). Some of its contributors even suffered from sustained attacks on social media. In personal communications between the research team and the website’s editors, the latter acknowledged online harassment as a serious challenge to authors and to the editors’ daily work. In light of this situation, the impact of toxic comments was an important focus of a survey of Conversation Canada authors who had published on either the English or French websites, conducted in late 2022.

Data collection and analysis

The current study employs a sequential mixed-methods approach that combines a quantitative survey and semi-structured interviews. The analysis of the survey and interview results is guided by the following research questions:

RQ1: How prevalent are toxic comments targeting Conversation Canada authors?

RQ2: What is the nature of toxic comments targeting Conversation Canada authors?

RQ3: How do Conversation Canada authors perceive and respond to toxic comments?

Our research team conducted a survey of Conversation authors’ perspectives on a variety of issues, such as their understanding of journalistic work, their experiences with the publishing process, and the impact of public writing on their scholarly work. The current paper only reports on data related to anti-press hostility. The development of survey questions involved consultation with Conversation Canada editors, and review of sample surveys from previous studies (e.g., Lewis et al., 2020; Löfgren Nilsson & Örnebring, 2016; Miller, 2021; Stahel & Schoen, 2020; Canadian Association of Journalists, 2021). The questionnaire and methodology was approved by the Research Ethics Board of Toronto Metropolitan University (file number 2022-138), and informed consent was obtained from participants. The Conversation Canada provided a list of its authors for survey distribution; however, it had no part in the collection of data or access to the raw data collected.

Dates of survey administration were September 7 to October 31, 2022. A total of 732 Conversation Canada authors responded to the survey, a 15% response rate. Following the survey, the research team conducted semi-structured interviews via Zoom with 37 respondents who agreed to further elaborate their survey answers, including some who had experiences of online harassment. Five Conversation Canada editors were also interviewed via Zoom, to contextualize established processes within the organization and, in some cases, to contextualize policies surrounding practice for avoiding and/or managing toxic comments. Emailed clarifications of interview comments were provided by some editors after interviews. Interviews were performed by one researcher or one of three research assistants.

Qualitative research methods should differ based on what suits a particular study (Rubin, 2021). The aim of interviews for this case study was to explore overall patterns that emerged in survey data, not directly compare an individual participant’s interview data to their survey data. Due to the specificity of subject matter being explored and to respect the time of participants, interviews were focused (Creswell & Poth, 2016), ranging from seven to 48 minutes in length and averaging 17 minutes in duration. In terms of value, there is “nothing magical or absolute” (Seidman, 2006, p. 20) about an interview’s timeframe. Longer interviews don’t always result in better data (Thorsteinson, 2017), and even a “one-shot question” (Patton, p. 378) can provide useful information. What’s more important is to recognize that, across a dataset, “the repetition of an aspect of experience . . . takes on weight” (Seidman, 2006, p. 127).

More than 200 authors expressed interest in being interviewed for this study by providing contact information in their survey responses. Researchers accessed that contact information through a randomized list of these participants, created  from  aggregated survey data versus individual surveys. Minimal curation of the list was performed to ensure there was no overrepresentation of authors from a particular university based on email-address extensions and that no participant appeared on both the French and English participant lists. One-hundred-ninety participants were contacted for interviews, but not all responded and not all who responded followed through with an interview. Researchers interviewed willing participants from the randomized list until no new themes were emerging and ~18% of survey respondents who expressed interest in being interviewed had participated. All interviews for this study were completed between October 13, 2022 and April 19, 2023. They were conversational in nature to allow for exploration of participants’ experience and opinions (Tracy, 2019), and to acknowledge there are “multiple realities constructed through lived . . . experiences” (Creswell & Poth, 2016, p. 81). However, the following sets of questions were prepared in advance and asked of each participant:

Questions for authors:

• How would you describe or identify the type of writing you do for The Conversation? (**would you describe your work as journalism, if not answered)
• What are the benefits of writing for The Conversation?
• What are the drawbacks of writing for The Conversation?

Questions for editors:

• Could you describe the editorial process of the publication of a Conversation article from pitch to posting?
• What are the challenges of working with academic authors to create journalistic work?
• What are the benefits of working with academic authors to create journalistic work?
• What do you consider to be explanatory journalism?

With a goal of contextualizing survey data, interview data were thematically analyzed (Riger & Sigurvinsdottir, 2016) by first using open coding to look for general themes throughout the data collection period, followed by focused coding (Rubin, 2021) once interviews were completed and themes were established. Participants shared information based on their particular perceptions, but individual quotes were selected to represent patterns seen in the overall dataset. Themes identified from interviews with Conversation authors included inconsistent definitions of journalism; positive experiences with publication; and, particular to this paper, negative experiences with publication that included a sub-theme of toxic interactions. These interactions included communication via website or social media comments, and/or communication sent via email, and ranged from questioning an author’s credibility to “personal attacks,” trolls, “rude comments,” and “bullying.” Themes for editors mirrored those of authors, but from a different perspective, with some specific observations related to challenging the norms of traditional gatekeeping processes.

FINDINGS

As shown in Table 1’s demographic statistics self-reported by survey respondents, although the survey focuses on Conversation Canada authors, the sample represents a diverse range of academics from a variety of backgrounds. In accordance with previous research (Lewis & Coddington, 2020; Miller, 2023a, 2023b; Waisbord, 2020), the analysis focuses on demographics of race/ethnicity, gender, disability, and religion, and assesses whether these markers make scholars writing in a journalistic style more prone to anti-press hostility.

For this purpose, the research team re-coded the demographic data in Table 1 as collapsed variables. Binary logistic regression and ordinal logistic regression were used to assess the potential influence of these variables on the frequency and severity of hostility experienced by the survey respondents.

 

Table 1. Survey Descriptive Statistics (Self-Reported by Respondents, Total Respondents = 732)

Respondents’ highest attained level of education:	Master’s (108), PhD (589), Other (35)
Respondents’ race: *	First Nations / Indigenous (5), Black (13), Latin (12), Middle Eastern (14), White (550), Asian (48), Mixed Race (25), Prefer not to answer (65)
Respondents’ gender identity: **	Woman (383), Man (275), Non-binary/other (24), Prefer not to answer (50)
Whether a respondent is identified as a person with a disability: ***	Yes (82), No (572), Prefer not to answer (78)

* Recoded as White = 0, non-White = 1
** Recoded as Man = 0, Woman = 1, Non-binary/other = 2
*** Recoded as No = 0, Yes = 1
Participants answering “prefer not to answer” are excluded from the regression analysis in Tables 6 to 8.

 

Table 2 presents the survey questions regarding toxic comments.

Analysis begins with an overview of the frequency and severity of anti-press and online hostility encountered by Conversation Canada authors. It then evaluates the role of social identity (e.g., gender, race, and religion) in triggering harassment. Further, it examines these markers’ influence on the actions taken in response to harassment and the perceived consequences from such incidents. The section ends by summarizing insights gleaned from the research team’s interviews with survey respondents that help contextualize survey data.

 

Table 2. Survey Questions

Have you received toxic comments online? * (Yes / No / Prefer not to answer)

How often do you receive toxic comments on your Conversation article(s)? ** (Never / Once / Several times over the past 12 months / At least once every month / At least once every week / Several times every week / Prefer not to answer)

In addition to the Conversation website, have you received toxic comments via other communication channels? Please select all that apply. (Online forums / Social media / Email / Posted mail / Not Applicable / Prefer not to answer)

What is the nature/content of the toxic comments you have received? Please select all that apply. (Skepticism of academic expertise / Appearance / Sexism / Racism / Sexual Orientation / Ideological / Not Applicable / Other / Prefer not to answer)

What is the subject of your article(s) that received toxic comments? Please select all that apply. (Podcasts / COVID-19 / Arts Business + Economy / Culture + Society / Education / Environment + Energy / Health (without COVID-19) / Politics / Science + Tech / Not Applicable / Prefer not to answer)

In terms of the impact of toxic comments, it happens that I:  Please select all that apply. (Avoid writing about specific issues because of the risk. / Avoid writing about specific persons/groups because of the risk. / Become afraid of the threats I receive. / Consider not expressing my viewpoint because of the threats I receive. / Chose to stop writing for The Conversation because of the comments I received. / Other / Not Applicable)

Overall, how would you assess the impact of toxic comments on your willingness to participate in knowledge mobilization initiatives outside of The Conversation? *** (They inspired me to work on more knowledge mobilization efforts. / There was no change in my knowledge mobilization efforts. / Neutral / They had minor negative impact on my knowledge mobilization efforts. / They had major negative impact on my knowledge mobilization efforts. / Not Applicable / Prefer not to answer)

* Recoded as No = 0, Yes = 1
** Recoded as Never/Not Applicable = 0, Once = 1, Several Times over Past Month = 2, At least once every month = 3, Several times every week = 4
*** Recoded as Inspired me to do more = -1, Not Applicable/Neutral/No Change = 0, Minor negative impact = 1, Major negative impact = 2
Participants answering “prefer not to answer” are excluded from the regression analysis in Tables 6 to 8.

 

Overall patterns of harassment targeting Conversation Canada authors

In the survey, a definition of a toxic comment was not provided to respondents. Instead, they were allowed to answer based on their own interpretations. Similarly, we did not define “knowledge mobilization” for respondents. However, as all were scholars, their understanding of this term would be influenced by Canada’s major research funding agencies. For example, the Social Sciences and Humanities Research Council of Canada (2023) defines knowledge mobilization as “an umbrella term encompassing a wide range of activities relating to the production and use of research results, including knowledge synthesis, dissemination, transfer, exchange, and co-creation or co-production by researchers and knowledge users” (para. 4). It is a broad term that covers a wide range of activities.

Table 3 shows the prevalence and severity of toxic comments targeting Conversation Canada authors. Among the 732 respondents to the survey, 27.5% (N = 201) reported having received toxic comments. 45.3% (91 out of 201) of these affected authors received at least several toxic comments within a 12-month period.

Toxic comments are not only confined to The Conversation Canada’s article comment section. Among the authors who were targeted, 55.7% (112 out of 201) reported receiving toxic comments on social media, while 39.3% (79 out of 201) mentioned they also received such comments through email. These are worrisome results considering The Conversation Canada’s smaller readership in comparison to more established news outlets.¹Noting that Conversation articles can be picked up by multiple outlets and, therefore, may have greater amplification than traditional news outlets in some cases. Based on survey data we cannot determine where all comments were received, e.g., website versus social media; however, the fact that Conversation readers have to register/cannot remain anonymous when commenting on the site and yet still leave toxic comments, according to editors and authors, is a further sign of the acceptability of such behavior in public discourse. Almost 27% (54 out of 201) of authors targeted by toxic comments reported that online hostility had a minor to significant negative impact on their willingness to participate in knowledge mobilization efforts. By contrast, a smaller portion of these authors, namely 12.9% (26 out of 201), found the hostility to be a catalyst, inspiring them to become more engaged in knowledge mobilization efforts.

 

Table 3: The Prevalence and Severity of Toxic Comments

Have you received toxic comments online?	Yes: 201 No: 521 Prefer not to answer: 10
Only those who answered “Yes” to the first question (N = 201) answered the subsequent questions.
How often do you receive toxic comments on your Conversation article(s)?	Once: 88 (43.8%) Several times over the past 12 months: 85 (42.3%) At least once every month to several times every week: 6 (3.0%) Never / Not Applicable / Prefer not to answer: 22 (10.9%)
In addition to The Conversation website, have you received toxic comments via other communication channels? (Multiple Choices)	Online forums: 36 (17.9%) Social media: 112 (55.7%) Email: 79 (39.3%) Posted mail: 11 (5.5%) Not applicable: 46 (22.9%)
Overall, how would you assess the impact of toxic comments on your willingness to participate in knowledge mobilization initiatives outside of The Conversation?	They inspired me to work on more knowledge mobilization efforts: 26 (12.9%) There was no change in my knowledge mobilization efforts: 105 (52.2%) Neutral: 15 (7.5%) They had minor negative impact on my knowledge mobilization efforts: 40 (19.9%) They had major negative impact on my knowledge mobilization efforts: 14 (7.0%) Not applicable / Prefer not to answer: 1 (0.5%)

 

Regarding the nature/content of toxic comments targeting Conversation Canada authors, Table 4 shows that the top three triggers are ideological disagreement (reported by 70.1%, 141 out of 201, of authors who received toxic comments), skepticism of academic expertise (47.3%, 95 out of 201), and gender discrimination (25.4%, 51 out of 201). In line with this trend, the top three types of Conversation articles targeted by toxic comments are politics (reported by 32.3%, 65 out of 201, of authors who received toxic comments), culture+ society (30.8%, 62 out of 201), and COVID-19 (26.9%, 35 out of 201). As both survey questions permitted multiple responses, the cumulative percentages in Table 4’s columns exceed 100%.

 

Table 4. The Nature and Content of Toxic Comments

Only those who answered “Yes” to the first question (N = 201) answered the subsequent questions.
What is the nature/content of the toxic comments you have received? (Multiple Choice)	Skepticism of academic expertise: 95 (47.3%) Appearance: 13 (6.5%) Sexism: 44 (21.9%) Racism: 33 (16.4%) Sexual orientation: 7 (3.5%) Ideological: 141 (70.1%) Other: 33 (16.4%)
What is the subject of your article(s) that received toxic comments? (Multiple Choice)	COVID-19: 35 (17.4%) Arts: 5 (2.5%) Business + Economy: 17 (8.5%) Culture + Society: 62 (30.8%) Education: 20 (10%) Environment + Energy: 19 (9.5%) Health (without COVID-19): 28 (13.9%) Politics: 65 (32.3%) Science + Technology: 18 (9.0%) Other: 39 (19.4%)

 

As evident in Table 5, the most common response to toxic comments among the survey respondents is “to consider not expressing my viewpoint because of the threats I receive” (19.9%, 40 out of 201), followed by “to avoid writing about specific issues because of the risk” (13.4%, 27 out of 201). Further examination of the qualitative responses given by respondents who selected the “other” option reveals negative emotions regarding online toxicity, such as anger, anxiety, contempt, and disappointment.

 

Table 5. Impact of Toxic Comments

Only those who answered “Yes” to the first question (N = 201) answered the subsequent questions.

In terms of the impact of toxic comments, it happens that I: (Multiple Choice)

Avoid writing about specific issues because of the risk: 27 (13.4%) Avoid writing about specific persons/groups because of the risk: 16 (8.0%) Become afraid of the threats I receive: 19 (9.5%) Consider not expressing my viewpoint because of the threats I receive: 40 (19.9%) Chose to stop writing for The Conversation because of the comments I received: 16 (8.0%) Other: 63 (31.3%) Not applicable: 70 (34.8%)

Harassment and social identity

To analyze whether social identity might predict Conversation Canada authors’ experience of online harassment, we conducted three regression analyses predicting the frequency and severity of toxic comments in relation to gender, race, and disability. A binary logistic regression analysis treating the three variables as predictors (Table 6) revealed that authors who are racialized (β=.669, p=.003) or people with disabilities (β=.665, p=.010) are more likely to encounter toxic comments. The Omnibus tests of model coefficients for this analysis rejected the null hypothesis (p<.001), indicating that at least one predictor variable is meaningfully related to the outcome variable. Additionally, the Hosmer and Lemeshow test for this analysis (p=.729) was consistent with the null hypothesis, indicating that the model fits the data well.

 

Table 6. Binary Logistic Regression Predicting Toxic Comments

Binary Logistic Regression	Race White = 0 Non-White = 1		Gender Man = 0 Woman = 1 Non-binary/other = 2		Disability No = 0 Yes = 1
	Coef. (β)	Sig.	Coef. (β)	Sig.	Coef. (β)	Sig.
Have you received toxic comments online? No = 0      Yes = 1	.669	.003*	.189	.256	.665	.010*

Omnibus tests of model coefficients p < .001
Hosmer and Lemeshow test p = .729
* p < .05

 

In subsequent analysis, the strongest predictor of toxic comments turned out to be race. An ordinal logistic regression analysis was conducted to investigate the relationship between the frequency of toxic comments and the three predictor variables, as shown in Table 7. The analysis found a statistically significant association between the frequency of toxic comments and “being White” (p=0.007). As indicated by the negative coefficient (β=−.609), being White is associated with experiencing a lower frequency of toxic comments. The analysis’ model fitting rejected the null hypothesis (p=.011), indicating that at least one of the predictors in the model is relevant. Additionally, its goodness-of-fit tests were consistent with the null hypothesis (Pearson p=.299, Deviance p=.751), indicating that there is no significant difference between the observed and expected frequencies, and the model fits the data well.

 

Table 7. Ordinal Logistic Regression Predicting Toxic Comments (I)

Ordinal Logistic Regression	How often do you receive toxic comments on your Conversation article(s)? No/Not Applicable = 0 Once = 1 Several times over past month = 2 At least once every month = 3 Several times per week = 4
	Coef. (β)	Sig.
Race
White = 0	−.609	.007*
Non-White = 1	0a	0
Gender
Man = 0	−.503	.269
Woman = 1	-.357	.449
Non-binary/other = 2	0a	0
Disability
No = 0	−.502	.060
Yes = 1	0a	0

* p < .05
Model fitting p = .011
Goodness-of-fit: Pearson p = .299, Deviance p = .751

Another ordinal logistic regression analysis investigating the relationship between the impact of toxic comments and the three predictor variables (Table 8) identified “Man = 0” as statistically significant (p=.042), and the negative coefficient (β=-1.293) suggested that being male is associated with a less significant effect of toxic comments on knowledge mobilization (when creating dummy variables, “Major negative impact” was coded as “2”). However, this finding needs to be interpreted with caution since the model fitting of this analysis failed to reject the null hypothesis (p = .285), and one of its goodness-of-fit tests (Pearson Chi-Square p = .004) also indicated a poor fit between the model and the data. Put differently, there is evidence to suggest men appear less likely to be affected by toxic comments in terms of their willingness to participate in knowledge mobilization. However, this finding should be interpreted cautiously as further research is required to confirm it.

 

Table 8. Ordinal Logistic Regression Predicting Toxic Comments (II)

Ordinal Logistic Regression	Overall, how would you assess the impact of toxic comments on your willingness to participate in knowledge mobilization initiatives outside of The Conversation? Inspired me to do more = -1 Not applicable/neutral/no change = 0 Minor negative impact = 1 Major negative impact = 2
	Coef. (β)	Sig.
Race
White = 0	−.131	.703
Non-White = 1	0a	0
Gender
Man = 0	−1.293	.042*
Woman = 1	-.909	.140
Non-binary/other = 2	0a	0
Disability
vNo = 0	.385	.360
Yes = 1	0a	0

* p < .05
Model fitting p = .285
Goodness-of-fit: Pearson p = .004, Deviance p = .071

 

Insights from semi-structured interviews

Interviews allowed for further contextualization of the experience of Conversation authors related to toxic interactions, and the subsequent impact. Eleven of 37 authors (all of whom had published on the English site) discussed receiving toxic comments. Authors who identified as being the target of toxicity largely pinpointed the subject of their articles, versus who they were, as the reason for being targeted.

Consistent with data gathered in the survey, many of the authors who had received toxic comments did not receive them directly on The Conversation website. As noted by one, “Every time I write for The Conversation I’ve got harassment emails, right. I have become really good with it. I have a spam folder” (Author 14, personal communication, March 8, 2023).

Having the comments turned off on articles that were deemed more likely to attract negative attention did not prevent toxic interactions:

There was a lot of bullying and the editor I worked with was really great, because, she said, like we can turn off comments at any time. But you know it was because we’re public-facing; even when you turn off the comments on The Conversation, my email is readily available. So I got a lot of yeah, just like Twitter messages, Facebook direct messages and emails that were really negative and a lot of hatred and bullying. (Author 25, personal communication, April 7, 2022, )

Two of five Conversation editors interviewed gave perspectives on managing toxic comments when asked about the experience of authors, noting that negative online interactions are common and increasingly unavoidable. For example:

More and more it’s like, especially our editors in the Culture and Society section where they’re dealing with race and Indigenous issues, especially if they’re first-time authors, the editors are talking to them to say you need to be prepared to, this is going to happen, you know. We will help when we can, but you need to acknowledge that it’s a risk, right?” (Editor 1, personal communication, October 21, 2022)

Ignoring/trying to avoid toxic comments was the most common strategy used by the authors we spoke with. However, several noted they were unsure how best to manage toxic interactions and thought there should be more guidance.

Others responded to negative comments with links to additional reports and papers to support the evidence cited in their articles. As described by one author: “They were personal . . . We just got roasted by some of the comments, but we gave as good as we got . . . quoted over 200 different citations [to support claims in the article]” (Author 6, personal communication, November 18, 2022).

Even for those who welcomed debate on their articles and who were comfortable engaging with harsh critics, things could go too far:

It’s not like as academics, we’re afraid to have tough conversations, and have new ideas kind of brought forth to us, but when attacks get personal, attacks get toxic and when they try to shut you down as a professional, which has happened to me, that’s where I think things sort of crossed the line.” (Author 2, personal communication, September 11, 2022)

However, most authors believed that toxic comments were to be expected in any online interaction. As noted by one, “This is, I think, part of the broader media system, but you can’t get away from it. It’s actually not something related to The Conversation. It’s just something that exists online” (Author 5, personal communication, November 17, 2022).

Some authors who had not received what they perceived to be toxic comments, and even a few who did, said they thought participating in dialogue on The Conversation site was worthwhile, and that it was beneficial to engage with the audience. Others found comments on the site to be better/more positive compared to other media organizations where they had published articles. An author who published on the French site described interactions with readers as “cordial” (Author 35, personal communication, June 3, 2023).

Editors noted a lingering sense of responsibility towards supporting authors and preventing toxic comments: “I do revisit stories to consider if there is anything that could have been done differently or something I might or should have addressed with the author” (Editor 4, personal communication, January 9, 2023). Many authors interviewed noted that more media training would be welcome. The Conversation’s policy at the time of this study was to provide none; according to interviews, it was considered the responsibility of cooperating universities to arrange such support for authors. However, the editor who revisited stories thought there was room to do more in relation to toxic interactions:

Likely, it would be a good direction for The Conversation to explore teaming up with researchers . . . to examine online harassment and university communities who have published stories with us–to offer a panel or, I have even thought, some kind of mutual mentoring support to authors involved in public intellectual life. (Editor 4, personal communication, January 9, 2023)

 

DISCUSSION AND CONCLUSION

In today’s increasingly polarized society, many public figures contend with an increasingly hostile online environment. In the case of journalists, previous research has shown that women, those who are racialized, and those covering contentious topics (e.g., structural inequality and racism) are more likely to be targeted by anti-press hostility (Miller, 2021a; Lewis et al., 2020; Waisbord, 2020, 2022a). In comparison, the experiences of academic authors writing for journalistic outlets have received less scholarly attention. These authors may not be “journalists” but they are writing in journalistic spaces where their presence and research is amplified to a wider audience using a journalistic style, under the banner of a “news” organization. To address this research gap, we investigated how Conversation Canada authors experience toxic comments using a combination of survey and interview data.

Our analysis paints a concerning picture. It suggests that toxic comments pose a notable challenge to The Conversation Canada’s journalism practice, as a sizable proportion of authors have encountered them and commenting norms have been changed to control them. The impact of these toxic comments is by no means trivial. According to survey data, they have a chilling effect, causing a number of the affected authors to avoid expressing their viewpoints (19.9%) or writing about certain topics (13.4%), as well as to reconsider their knowledge mobilization efforts (26.9%), thereby fostering an atmosphere of self-censorship. Such findings support news industry demands for serious efforts to curb anti-press hostility (Fenlon, 2022). While this research was under review, The Conversation Canada held a workshop at the annual gathering of the Canadian Federation for the Humanities and Social Sciences, focusing on strategies for dealing with online harassment. Discussions during the workshop emphasized peer support and collective advocacy to encourage universities to establish policies that support faculty members in handling online hostility.

RQ1 focused on the frequency and severity of toxic comments directed at Conversation Canada authors. As reported in the previous section, 27.5% of survey respondents have encountered toxic comments. Among the respondents who experienced harassment, 45.3% reported receiving multiple toxic comments within a 12-month period. Meanwhile, 26.9% of the respondents who experienced harassment considered toxic comments as having a negative impact on their enthusiasm for knowledge mobilization.

At first glance, these results suggest that hostility directed at Conversation Canada authors may not be as severe or frequent as that directed at professional journalists surveyed in previous research (e.g., Lewis et al., 2020; Lofgren Nilsson & Ornebring, 2016; Miller, 2021; Stahel & Schoen, 2020). However, two factors must be considered. First, as many academics contribute to websites like The Conversation Canada out of a passion for knowledge mobilization rather than a job necessity, toxic comments can prompt them to reconsider the priority of public writing in their work routines, negatively impacting public discourse.

Second, as mentioned previously, The Conversation Canada’s exclusive focus on articles supported by academic evidence is meant to attract readers interested in engaging in constructive conversations with scholars. For this purpose, the website requires readers to register with an email address or other online credentials before posting a comment. Despite this preventative measure, many authors continue to be the target of toxic comments, as evidenced by the findings. As of May 2025, The Conversation Canada website maintains the policy of closing the comment section for selected articles, underscoring the fact that toxic comments remain a major operational challenge.

RQ2 investigated the triggering factors for toxic comments. The survey explored the relationship between three visible markers of social identity (gender, race, and disability) and the frequency of toxic comments, as well as Conversation Canada authors’ responses. According to the regression analysis results, race is a significant variable in predicting the frequency of receiving toxic comments, followed by disability (Tables 6 & 7). Compared with many previous studies that emphasized the disproportionate impact of online harassment on women journalists (e.g., Chen et al., 2020; Miller & Lewis, 2022; Tandoc et al., 2021), the current analysis found no statistically significant correlation between gender and the predicted frequency of receiving toxic comments. This could support authors’ perceptions that, in some cases, it is the subject of articles that is more likely to attract toxic comments. This could also, perhaps, be unique to The Conversation’s environment, which some authors noted was comparatively more civil than other publishing environments.

Table  8  shows  a  statistically  significant correlation between gender and responses to toxic comments; however, this finding requires cautious interpretation, given the unsatisfactory model fitting and goodness-of-fit results. While male authors were less likely to change their behavior in response to online harassment, echoing previous findings in studies such as Miller (2023b), and Stahel and Schoen (2020), women were more likely to change behavior to avoid such interactions. This is not unexpected, given that comments targeted at women are often severe, violent, and personal (Poestti & Bonchetva, 2022). However, it is greatly concerning in terms of mitigating the ability of women academics/journalists to interact with the public (which, as noted previously, might impact career opportunities) and limiting the voices of women in online discourse, in general. This is another area requiring future research.

How should these findings be interpreted? The content of toxic comments and the specific articles they target, as reported by survey respondents, provide important clues. The top three triggers for toxic comments are ideological disagreement, skepticism towards academic expertise, and gender discrimination. The Conversation Canada sections most frequently targeted by toxic comments are politics, culture and society, and COVID-19. Overall, these results suggest the majority of toxic comments directed at Conversation Canada authors are politically motivated attacks on their expertise and standpoints, which, according to Miller’s (2023b) classification of various forms of harassment, belong to “incivility and disruptive harassment” that attacks the authors’ identity as public intellectuals (p. 200).

Accordingly, the anti-press and online hostility targeting The Conversation Canada is indicative of mob censorship (Waisbord, 2020), which results from the combination of three developments: “easy public access to public figures (e.g., journalists and academics), the presence of toxic internet right-wing and far-right cultures, and populist demonization of the mainstream press” (p. 1037). Mob censorship exemplifies a worrying trend on the internet, namely that intensifying populism and political polarization lead to an upsurge in the prevalence of harmful discourses over constructive ones. This situation underscores the importance of restoring the collective well-being and democratic health on digital platforms.

RQ3 examined the responses of Conversation Canada authors to toxic interactions. It is concerning to discover that 26.9% of authors who have experienced online harassment viewed toxic comments as having a negative impact on their interest in sharing knowledge. Nonetheless, based on interviews, neither The Conversation Canada nor the authors’ affiliated universities provided adequate organizational support for dealing with toxic comments at the time of data collection. Often, authors were left on their own to figure out how best to manage comments, particularly those arriving outside of the website. Based on interviews, these efforts required significant time and emotional labour. Given the growing demand from funding agencies for researchers to engage in knowledge mobilization activities, universities must urgently develop resources and policies that support scholars who seek psychological or legal assistance due to experiencing severe online harassment that can be amplified because of their visibility in news publications with resulting exposure to a wider audience. Existing resources that identify best practices for dealing with online harassment in a journalism context could be adapted for this purpose (e.g., Bedei, 2024; McGregor, Vilk, & Mohamed, 2024), and such adaptation would be a valuable area of future research.

In summation, this study promotes a more nuanced understanding of anti-press hostility, demonstrating that even those who occasionally contribute to journalistic outlets are subject to it. The rise of mob censorship as a significant obstacle to the daily operations of media startups such as The Conversation Canada is alarming, as such independent media rely on academics and other non-journalists for content but lack the means to protect them from anti-press hostility. The study’s findings also echo recent research on digital media criticism (Carlson et al., 2021; Cheruiyot, 2022; Figenschou & Ihlebæk, 2019), showing that toxic comments, driven by cynicism towards journalism and authority, actively attempt to undermine the epistemic legitimacy of journalism and question its social function.

Therefore, it is crucial that governments, digital platforms, news media, and other stakeholders work together to shield the press, and those contributing to the press who may not be “journalists,” from digital hostility. “Only if mob censorship is recognized across the board as a collective, social problem, rather than a series of unfortunate events affecting individual reporters, it would [sic (would it)] be possible to envision effective responses to online violence” (Waisbord, 2023, p. 1767).

The prevalence of cross-platform toxicity underscores the need for a balanced approach to content moderation. While the decision of The Conversation Canada to close the comment section for certain articles is understandable, it may result in the loss of the opportunity to contain negative interactions within a specific environment, which increases the overall burden of harassment experienced by academics elsewhere. Thus, it is essential to consider how comment policies can impact not only the immediate discourse on news websites but also the broader implications for authors’ experiences online.

This study has limitations. Due to the focus of the survey and interviews on the experiences of Conversation Canada authors, the findings cannot be generalized to countries with different media cultures or all inhabitants of the field of journalism. We also recognize that some authors received toxic comments that were unrelated to their race, gender, or disabilities. The regression analyses presented in Tables 6 to 8 reveal limited explanatory power, suggesting that the interplay of toxic comments and social identity markers does not fully account for digital hostility. In addition, the study gave limited consideration to the coping strategies employed by Conversation Canada authors in the face of mob censorship. In light of these limitations, future research may adopt a comparative design to investigate how academics experience anti-press and online hostility in different countries or conduct in-depth, broader interviews to determine how academics at different career stages view the benefits and drawbacks of writing for the media. For example, are pre-tenure or contract faculty more likely to risk harassment for the perceived benefit of career advancement?

 

ENDNOTES

1. Noting that Conversation articles can be picked up by multiple outlets and, therefore, may have greater amplification than traditional news outlets in some cases.

 

---

Dr. Sibo Chen is an Associate Professor of Professional Communication at Toronto Metropolitan University. His research explores how political polarization is communicated in the public sphere, focusing on three topics: political contention over climate change, online mis/disinformation, and the rise of anti-Asian racism. He is a leader in the SSHRC-funded Explanatory Journalism project. Email: sibo.chen@torontomu.ca

Dr. Nicole Blanchett is an associate professor of Journalism at Toronto Metropolitan University, researching journalistic roles and the impact of technology on practice. A former television journalist, she heads the Canadian team of the Journalistic Role Performance project, is a leader in the SSHRC funded Explanatory Journalism project, co-director of the Local News Research Project, and is part of Canada’s Worlds of Journalism Study. Email: nicole.blanchett@torontomu.ca

S. Lecourt author info withheld by request.

---

Acknowledgement

The authors would like to acknowledge the assistance of Anthony Milton, Macenzie Rebelo, and Racy Rafique in the data gathering process.

Disclosure Note

Nicole Blanchett is on the editorial board of Facts & Frictions; however, this paper was double-blind peer-reviewed and her role as an author was not known by other members of the board or designated reviewers at any point during the submission, review, and approval process.

Author Contributions

Sibo Chen and Nicole Blanchett conceived and designed this study, while working in a larger research team for a broader project. The original dataset was collected by the Explanatory Journalism project at Toronto Metropolitan University. Sibo Chen conducted the statistical analysis, while Nicole Blanchett led data collection for interviews with selected survey respondents, analysis of interview data, and wrote relevant content related to qualitative data in the manuscript. Sibo Chen drafted the rest of the initial manuscript, and the subsequent revisions of the manuscript were completed collectively by Sibo Chen and Nicole Blanchett. S. Lecourt conducted the study’s French interviews and translated them into English, and assisted with preliminary analysis of qualitative data.

Data Availability

The datasets generated during and/or analyzed during the current study are not publicly available due to research ethics requirements but are available from the corresponding author on reasonable request.

Ethical Approval

The questionnaire and methodology for this study was approved by the Research Ethics Board of Toronto Metropolitan University (file number 2022-138).

Informed Consent

Informed consent was obtained from all survey participants.

---

REFERENCES

Allen, G., & Lucky, N. (2023). (New) media and the circulation of knowledge: A historical framework for The Conversation Canada. Information & Culture, 58(3), 221-246.

Barão da Silva, G., Sbaraini Fontes, G., & Marques, F. P. (2022). Risks and resilience in the case of Brazilian female journalists: How women perceive violence against media professionals and cope with its effects. Journalism Studies, ahead-of-print. https://doi.org/10.1080/1461670X.2022.2150873

Bedei, C. (2024, December 16). How to deal with trolls: Advice for journalists targeted with online harassment. https://web.archive.org/web/20250518172241/ https://ijnet.org/en/story/how-deal-trolls-advice-journalists-targeted-online-harassment

Blanchett, N. (2021). Participative gatekeeping: The intersection of news, audience data, newsworkers, and economics. Digital Journalism, 9(6), 773-791.

Blanchett, N., Brin, C., & Duncan, S. (2024). Catch, Engage, Retain: Audience-Oriented Journalistic Role Performance in Canada. Journalism Practice, 18(9), 2258-2280.

Blanchett, N., Ono, S., Lecourt, S., & Hillis, C. (2025). The Conversation Canada and journalistic boundaries: A case study in the challenges of defining journalism. Canadian Journal of Communication, 50(1), 115-144.

Brin, C., & Charlton, S. (2022). Canada. Reuters institute digital news report 2022. Reuters Institute. https://web.archive.org/web/20250518172339/https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2022/canada

Broockman, D., & Kalla, J. (2022). The manifold effects of partisan media on viewers’ beliefs and attitudes: A field experiment with Fox News viewers. OSF Preprints. https://doi.org/10.31219/osf.io/jrw26

Canada Press Freedom Project (2024). 2023 Annual Report. https://web.archive.org/web/20250518172508/https://canadapressfreedom.ca/the-canada-press-freedom-project-2023-report/

The Canadian Association of Journalists (2021). Canadian newsroom diversity survey (2021).  https://web.archive.org/web/20250518174829/https://caj.ca/programs/diversity-survey/survey-results-2021/

The Conversation United States (2022). 2021 annual report: Expert knowledge for the public good. https://cdn.theconversation.com/static_files/files/1813/The_Conversation_US_2021_Annual_Report.pdf

The Conversation United States (2024). 2023 annual report: Making sense of a complex world. https://cdn.theconversation.com/static_files/files/2935/Impact_Report_2023-_The_Conversation_US.pdf

Carlson, M., Robinson, S., & Lewis, S. C. (2021). Digital press criticism: The symbolic dimensions of Donald Trump’s assault on U.S. journalists as the “enemy of the people”. Digital Journalism, 9(6), 737–754. https://doi.org/10.1080/21670811.2020.1836981

Chen, G. M., Pain, P., Chen, V. Y., Mekelburg, M., Springer, N., & Troger, F. (2020). ‘You really have to have a thick skin’: A cross-cultural perspective on how online harassment influences female journalists. Journalism, 21(7), 877-895. https://doi.org/10.1177/1464884918768500

Chen, S. (2022). Reporting in a time of crisis: Progressive alternative media’s coverage of the COVID-19 pandemic in Canada. Journalism Practices, 18(6), 1542-1559. https://doi.org/10.1080/17512786.2022.2090999

Cheruiyot, D. (2022). Comparing risks to journalism: Media criticism in the digital hate. Digital Journalism, 12(3), 294–313. https://doi.org/10.1080/21670811.2022.2030243

COVID scientists in the public eye need protection from threats (2021, October 13). Nature (London), 598(7880), 236–236. https://doi.org/10.1038/d41586-021-02757-3

Creswell, J. W., & Poth, C. N. (2016). Qualitative inquiry and research design: Choosing among five approaches. Sage publications.

Davis Kempton, S., & Connolly-Ahern, C. (2022). “Who’s going to be a creep today?” Understanding the social media experiences of women broadcast journalists. Social Media + Society, 8(2), Article 205630512211084. https://doi.org/10.1177/20563051221108410

Deavours, D., Heath, W., Miller, K., Viehouser, M., Palacios-Plugge, S., & Broussard, R. (2022). Reciprocal journalism’s double-edged sword: How journalists resolve cognitive dissonance after experiencing harassment from audiences on social media. Journalism, ahead-of-print, article 146488492211096. https://doi.org/10.1177/14648849221109654

Deuze, M., & Witschge, T. (2018). Beyond journalism: Theorizing the transformation of journalism. Digital Journalism, 19(2), 165–181. https://doi.org/10.1177/1464884916688550

Eschner, K. (2022). Poisoned well: Industry roundtable on journalists and online hate. Canadian Association of Journalists & Canadian Journalism Foundation. https://cjf-fjc.ca/poisoned-well-industry-roundtable-on-journalists-and-online-hate/

Fadnes, I., Krøvel, R., & Orgeret, K. S. (2019). Where are we now? – (Almost) a decade with The UN Plan of Action on the Safety of Journalists and the Issue of Impunity. Conflict & Communication Online, 18(1), 1–6.

Fenlon, B. (2022, February 08). Our journalists are facing more harassment, threats for doing their jobs. CBC News. https://web.archive.org/web/20250518172747/https://www.cbc.ca/news/editorsblog/editor-note-pandemic-protests-media-experience-1.6343672

Figenschou, T. U., & Ihlebæk, K. A. (2019). Media criticism from the far-right: Attacking from many angles. Journalism Practice, 13(8), 901–905. https://doi.org/10.1080/17512786.2019.1647112

Frangou, C. (2019, October 23). The growing problem of online harassment in academe. University Affairs. https://web.archive.org/web/20250518172858/https://universityaffairs.ca/features/the-growing-problem-of-online-harassment-in-academe/

Gewin, V. (2018). Real-life stories of online harassment–and how scientists got through it. Nature (London), 562(7727), 449–450. https://doi.org/10.1038/d41586-018-07046-0

Glassey, M. (2017, August 9). How (and why) we lock accounts at The Conversation US. https://web.archive. org/web/20250518172952/https://theconversation. com/how-and-why-we-lock-accounts-at-the-conversation-82211

Gosse, C., Veletsianos, G., Hodson, J., Houlden, S., Dousay, A., Lowenthal, P. R., & Hall, N. (2021). The hidden costs of connectivity: nature and effects of scholars’ online harassment. Learning, Media and Technology, 46(3), 264–280. https://doi.org/10.1080/17439884.2021.1878218

Hermida, A., & Young, M. L. (2019). From peripheral to integral? A digital-born journalism not for profit in a time of crises. Media and Communication 7(4): 92–102, https://doi.org/10.17645/mac.v7i4.2269

Hiltunen, I. (2022). External interference in a hybrid media environment. Journalism Practice, 16(10), 2106–2124. https://doi.org/10.1080/17512786.2021.1905539

Holton, A. E., Bélair-Gagnon, V., Bossio, D., & Molyneux, L. (2023). “Not their fault, but their problem”: Organizational responses to the online harassment of journalists. Journalism Practice, 17(4), 859–874. https://doi.org/10.1080/17512786.2021.1946417

Kim, C., & Shin, W. (2022). Harassment of journalists and its aftermath: Anti-press violence, psychological suffering, and an internal chilling effect. Digital Journalism, ahead-of-print. https://doi.org/10.1080/21670811.2022.2034027

Lewis, S. C., Zamith, R., & Coddington, M. (2020). Online harassment and its implications for the journalist- audience relationship. Digital Journalism, 8(8), 1047-1067. https://doi.org/10.1080/21670811.2020.1811743

MacDonald, S., & Kalentsis, A. (2023, July 1). The stabbing attack at the University of Waterloo underscores the dangers of polarizing rhetoric about gender. The Conversation Canada. https://web.archive.org/web/20250518173259/https://theconversation.com/the-stabbing-attack-at-the-university-of-waterloo-underscores-the-dangers-of-polarizing-rhetoric-about-gender-208904

Malcorps, S., Libert, M., & Le Cam, F. (2022). A matter of organisational silence: Media managers struggling to make sense of the online harassment of journalists as a collective issue in journalism. Digital Journalism, 11(10), 1868–1885. https://doi.org/10.1080/21670811.2022.2140301

Mann, T. E. (2016, February 29). Explanatory journalism: A tool in the war against polarization and dysfunction. https://www.brookings.edu/blog/fixgov/2016/02/29/explanatory-journalism-a-toolin-the-war-against-polarization-and-dysfunction

Miller, K. C. (2021). Harassment’s toll on democracy: The effects of harassment towards US journalists. Journalism Practice, 17(8), 1607–1626. https://doi.org/10.1080/17512786.2021.2008809

Miller, K. C. (2023a) Hostility toward the press: A synthesis of terms, research, and future directions in examining harassment of journalists. Digital Journalism, 11(7), 1230-1249. https://doi.org/10.1080/21670811.2021.1991824

Miller, K. C. (2023b). The “price you pay” and the “badge of honor”: Journalists, gender, and harassment. Journalism & Mass Communication Quarterly, 100(1), 193–213. https://doi.org/10.1177/10776990221088761

Miller, K. C., & Kocan, S. (2022). Taking to the streets: The effects of in-the-field harassment against journalists covering protests. Journalism, 25(2), 275-294. https://doi.org/10.1177/14648849221138429

Miller, K. C., & Lewis, S. C. (2022). Journalists, harassment, and emotional labor: The case of women in on-air roles at US local television stations. Journalism, 23(1), 79–97. https://doi.org/10.1177/1464884919899016

Löfgren Nilsson, M., & Örnebring, H. (2016). Journalism under threat: Intimidation and harassment of Swedish journalists. Journalism Practice, 10(7), 880–890. https://doi.org/10.1080/17512786.2016.1164614

McGregor, S., Vilk, V., & Mohamed, J. (2024) The power of peer support: Helping journalists persevere in the face of online abuse. Pen America. (2024, April 3). Pen America. https://web.archive.org/web/20250518175054/https://pen.org/report/peer-support/

Nelson, J. L. (2022). “Worse than the harassment itself.” Journalists’ reactions to newsroom social media policies. Digital Journalism, 11(8), 1456–1474. https://doi.org/10.1080/21670811.2022.2153072

Nogrady, B. (2024). Harassment of scientists is surging–institutions aren’t sure how to help. Nature (London), 629(8013), 748–750. https://doi.org/10.1038/d41586-024-01468-9

Obermaier, M., Hofbauer, M., & Reinemann, C. (2018). Journalists as targets of hate speech. How German journalists perceive the consequences for themselves and how they cope with it. Studies in Communication and Media, 7(4), 499–524. https://doi.org/10.5771/2192-4007-2018-4-499

Oksanen, A., Celuch, M., Latikka, R., Oksa, R., & Savela, (2022). Hate and harassment in academia: The rising concern of the online environment. Higher Education, 84(3), 541–567. https://doi.org/10.1007/s10734-021-00787-4

Paletz, S. B., Johns, M. A., Murauskaite, E. E., Golonka, E. M., Pandža, N. B., Rytting, C. A., … & Ellis, D. (2023). Emotional content and sharing on Facebook: A theory cage match. Science Advances, 9(39), eade9231.

Pearson, M., Seglins, D., & Lindeman, T. (2022). Taking Care: A report on mental health, well-being and trauma among Canadian media workers. The Canadian Journalism Forum. https://web.archive.org/web/20250518174219/https://www.journalismforum.ca/taking-care-report

Posetti, J., Shabbir, N., Maynard, D., Bontcheva, K., & Aboulez, N. (2021). The chilling: Global trends in online violence against women journalists; research discussion paper. UNESCO. https://web.archive.org/web/20250518174357/https://unesdoc.unesco.org/ark:/48223/pf0000377223

Quandt, T. (2018). Dark participation. Media and Communication, 6(4), 36–48. https://doi.org/10.17645/mac.v6i4.1519

Reporters without Borders (2023). 2023 world press freedom index. https://web.archive.org/web/20250518174513/https://rsf.org/en/2023-world- press-freedom-index-journalism-threatened-fake-content-industry

Seidman, I. (2006). Interviewing as qualitative research: A guide for researchers in education and the social sciences. Teachers College.

Social Sciences and Humanities Research Council of Canada (2023). Guidelines for effective knowledge mobilization. https://web.archive.org/web/20250518174553/https://www.sshrc-crsh.gc.ca/funding-financement/policies-politiques/knowledge_mobilisation-mobilisation_des_connaissances-eng.aspx

Stahel, L., & Schoen, C. (2020). Female journalists under attack? Explaining gender differences in reactions to audiences’ attacks. New Media & Society, 22(10), 1849– 1867. https://doi.org/10.1177/146144481988533

Tandoc, E. C., Sagun, K. K., & Alvarez, K. P. (2021). The digitization of harassment: Women journalists’ experiences with online harassment in the Philippines. Journalism Practice, 17(6), 1198–1213. https://doi.org/10.1080/17512786.2021.1981774

Thorsteinson, T. J. (2017). A meta-analysis of interview length on reliability and validity. Journal of Occupational and Organizational Psychology, 91(1), 1–32. https://doi.org/10.1111/joop.12186

Varano, L. (2024, January 1). Year in review: Our Top 10 most read stories. The Conversation Canada (Newsletter). https://web.archive.org/web/20250518175129/https://theconversationcanada.createsend1.com/t/r-e-titiajd-bjrydidulu-r/

Waisbord, S. (2020). Mob censorship: Online harassment of US journalists in times of digital hate and populism. Digital Journalism, 8(8), 1030-1046. https://doi.org/10.1080/21670811.2020.1818111

Waisbord, S. (2022a). Can journalists be safe in a violent world? Journalism Practice, 16(9), 1948–1954. https://doi.org/10.1080/17512786.2022.2098524

Waisbord, S. (2022b). Trolling journalists and the risks of digital publicity. Journalism Practice, 16(5), 984–1000. https://doi.org/10.1080/17512786.2020.1827450

Waisbord, S. (2023). Mob censorship revisited: Questions, findings, and challenges. Digital Journalism, 11(10), 1761–1768. https://doi.org/10.1080/21670811.2023.2214905

Westlund, O. (2021). Advancing research into dark participation. Media and Communication, 9(1), 209–214. https://doi.org/10.17645/mac.v9i1.1770

Westlund, O., Krøvel, R., & Skare Orgeret, K. (2022). Newsafety: infrastructures, practices and consequences. Journalism Practice, 16(9), 1811–1828. https://doi.org/10.1080/17512786.2022.2130818

World Economic Forum (2024, January 10). Global risks 2024: At a turning point. https://www.weforum.org/publications/global-risks-report-2024/in-full/global-risks-2024-at-a-turning-point/#global-risks-2024-at-a-turning-point

Wright, J. M., Chun, W. H. K., Clarke, A., Herder, M., & Ramos, H. (2022). Protecting expert advice for the public: promoting safety and improved communications. Facets (Ottawa), 7, 482–508. https://doi.org/10.1139/facets-2021-0181