An examination of psychological distress and moral injury in journalists exposed to online harassment

ABSTRACT

Background: Studies show that journalists face repeated, intense online harassment. While data reveal this is distressing to the profession, no detailed psychological study has been undertaken defining what this distress entails.

Objective: To undertake a descriptive study examining the emotional wellbeing of journalists subject to online harassment.

Method: Data from 246 journalists working for a Canadian news organization were collected via a secure study website. Information collected included demographics, harassment metrics (frequency and reasons for harassment); level of organizational support rated on a simple analog scale (0–10, with low scores indicating poor support); psychometric symptoms (anxiety: GAD-7; depression: PHQ-9; posttraumatic stress disorder: PCL-5; moral injury: Toronto Moral Injury Scale for Journalists).

Results: The mean age of the sample was 43.07 (SD = 11.83) years. Fifty per cent were female. Harassment occurred at least weekly in 65 (26%) of the sample. Anxiety scores in the moderate to severe range were reported by 74 (30.1%) journalists while 34 (13.8%) had PTSD symptoms above the PCL-5 threshold for potential PTSD. Frequency of harassment correlated significantly with anxiety (r = 0.16, p = .014), depression (r = 0.15, p = .022), PTSD (r = 0.2, p = .002) and moral injury (r = 0.3, p < .002). Moral injury correlated significantly with anxiety (r = .40), depression (r = .41) and PTSD (r = .44; Spearman’s rho, p = .001 for all) scores. Organization support was rated as modest (M = 5.80, SD = 3.01).

Conclusions: Frequency of online harassment is associated with a range of emotional responses. While anxiety is the predominant emotion, clinically significant symptoms of PTSD affect a substantial minority of journalists. Moral injury is linked to other indices of emotional distress. News organizations should do more to address the challenges posed by harassment and better support their journalists.

HIGHLIGHTS

• Online harassment of journalists.

• Psychological distress including anxiety, depression and PTSD.

• Moral injury.

1. Introduction

There is a burgeoning literature on how journalists may be affected psychologically by their work. Studies have examined the effects of war (Feinstein, 2013; Osmann, Dvorkin, et al., 2021), government suppression (Abhishek, 2022; Harlow et al., 2023), drug-related violence (Feinstein, 2013), crime reporting (Feinstein, 2012; Petersen & Soundararajan, 2020), the COVID pandemic (Osmann, Selva, et al., 2021) and natural disasters (Buchanan & Keats, 2011; Dworznik-Hoak, 2020) on the emotional health of the Fourth Estate. Journalists are at risk for an array of psychiatric difficulties including depressive and anxiety disorders and posttraumatic stress disorder (Osmann, Dvorkin, et al., 2021). More recently, data have shown that moral injury may also arise in response to work-related stressors in the profession (Feinstein et al., 2018). One such stressor is online harassment which emerged as a key morally injurious event in the development of the Toronto Moral Injury Scale for Journalists (Osmann et al., 2024). Findings from a broader trauma literature reveal that moral injury, which refers to a condition that may arise from witnessing, perpetrating, or failing to prevent acts that transgress their moral compass (Litz et al., 2009) is in turn associated with posttraumatic stress disorder (Currier et al., 2019; Levi-Belz et al., 2020).

Much has been written about a relatively new source of stress that journalists must deal with, namely online harassment (Björkenfeldt, 2023). The evidence shows this is global as attested to by studies from North (Holton et al., 2023) and South (Dodds et al., 2024) America, Europe (Sampaio-Dias et al., 2023), Africa (Uwalaka & Amadi, 2023), Australia (Jane, 2018) and Asia (Tandoc et al., 2023). Given that the work of journalists is often in the public domain, they are particularly vulnerable to being targeted in this fashion across multiple online platforms like email, X (formerly Twitter), and Facebook, amongst others and by blogging, doxing and website postings too (Hiltunen, 2019; Jamil, 2020). Online harassment has now been highlighted as one more danger journalists must cope with alongside longer-standing risks like assault, arrests and public denigration (Reporters Without Borders, 2020). The rapidity with which this hostile environment is evolving in striking. Journalists in the USA and Canada now consider online harassment as one of their leading threats (Westcott, 2019).

No study to date has undertaken an in-depth assessment of how harassment is affecting the psychological health of journalists. There is a consensus within an increasingly large literature that the effects of harassment are distressing (Björkenfeldt, 2023; Celuch et al., 2023; Kim & Shin, 2025; Lee & Park, 2024; Posetti & Shabbir, 2022), but what this distress entails in journalists has not been systematically investigated. Clues may be gleaned elsewhere. For example, a systematic review of cyber harassment and mental health that included 43 articles from diverse groups (university students, faculty, politicians, the general population) revealed strong associations between being targeted and symptoms such as depression, anxiety, suicidal ideation and panic (Stevens et al., 2021). While these cross sectional studies were not designed to establish causality, the associations derived at should be seen in the context of harassment that entailed being threatened with death and physical and sexual violence, having one’s face photo-shopped onto another person’s body, having a profile set up in the victim’s name on a pornography website and having intimate images posted online.

The absence of data on the mental health of journalists has provided the impetus for the current research. The aims of the study are therefore two-fold: To determine the rates of anxiety, depression and PTSD symptoms in a representative sample of active journalists and to determine the strength of their association with moral injury. While the analysis was largely exploratory, we hypothesized that the frequency of harassment would be associated with more symptoms of psychopathology, and that women would be more symptomatic than men in keeping with the general psychiatry literature relating to depression, anxiety and PTSD (Kuehner, 2017; Perrin et al., 2014).

2. Method

Journalists working for a Canadian news organization were alerted to the study by an internal organization email. This generated a group of 246 participants whose study data were collected with Qualtrics, a secure online data collection tool licensed by the principal investigator’s host institution. Journalists were provided with a password and a link to enter the site. Once logged on, they were first taken to a description of the study and the consent form. Clicking on a button at the end of the form signalled consent and automatically took participants to the first of six questionnaires. The data collected were anonymized.

Data collected include the following:

1. Demographic information: age, sex, gender, marital status, number of years worked as a journalist, and type of journalism.

2. Harassment data were assessed with a multiple choice format and covered the following questions: platform for harassment (email, Facebook, Instagram, X [formerly Twitter], and WhatsApp); frequency of harassment; perceived reasons for being targeted; nature of the harassment; whether harassment was becoming more frequent and vicious; whether psychological support was provided in response to harassment and how journalists perceived the level of support provided by the news organization in relation to harassment. This was rated on a simple analog scale of 0–10 with low scores indicating poor support.

3. Psychological data: past contact with a mental health specialist; reason for contact with a mental health specialist; history of receiving treatment for psychological trauma; weekly units of alcohol divided according to sex given the Canadian Medical Association guidelines that stipulate a maximum of 14 units of alcohol per week for men and 9 units per week for women (Bondy et al., 1999). A unit refers to a shot of spirits, bottle of beer, or glass of wine.

All the psychometric measures chosen have been widely used, including in previous journalism research (Osmann, Selva, et al., 2021).

1. Anxiety symptoms over the past two weeks were recorded with the Generalized Anxiety Disorder Scale (GAD-7; Spitzer et al., 2006) that contains seven questions scored on a 4-point Likert scale (0 = not at all, 1 = several days, 2 = more than half the days, 3 = nearly every day). The GAD-7 is stratified into four levels of symptom severity: minimal, mild, moderate, and severe anxiety. The Cronbach’s α for our data was 0.93.

2. Depressive symptoms over the past two weeks were documented with the Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001) that contains nine questions scored on a 4-point Likert scale (0 = not at all; 1 = several days; 2 = more than half the days; 4 = nearly every day). The PHQ-9 is stratified into five levels of symptom severity: none-minimal, mild, moderate, moderately severe, and severe depression. The Cronbach α for our data was 0.89.

3. Symptoms of PTSD over the past month were noted with the PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013), a 20-item scale scored on a 5-point Likert scale (0 = not at all; 1 = a little bit, 2 = moderately, 3 = quite a bit; 4 = extremely). For the stressor criterion, we adhered to the PCL-5 wording as follows: ‘Below is a list of problems that people sometimes have in response to a very stressful experience. Keeping your worst event in mind, please read each problem carefully and then select one of the numbers to the right to indicate how much you have been bothered by that problem in the past month’. Journalists were asked to specifically choose a work-related event or events that qualified as their ‘very stressful experience/worst event.’ A total score of greater than or equal to 31–33 is indicative of potential PTSD (National Center for PTSD, 2025). The PCL-5 has four subscales: intrusion, avoidance, negative thoughts, and autonomic hyperarousal. One can also derive a tentative diagnosis of PTSD from these subscale scores as follows: One score of 2 or more from the intrusion and avoidance subscales plus two scores of 2 or more from the negative thoughts and arousal subscales. To ascertain the likelihood of a journalist having PTSD, we took a conservative approach; only those journalists who met both thresholds were deemed potentially to have the disorder. The Cronbach α for our data was 0.96.

4. Symptoms of Moral Injury associated with online harassment (no time period specified) were elicited with the Toronto Moral Injury Scale for Journalists (TMIS-J; Osmann et al., 2024), scored on a 5-item Likert scale (0 = none; 1 = minimal; 2 = moderate; 3 = quite a lot; 4 = severe). For the morally injurious event, journalists were asked to choose a specific work related event or events. The Cronbachα for our data was 0.86.

2.1. Statistical analyses

Data were analysed with IBM SPSS Statistics version 29. There were no missing psychometric data as the method of website data collection did not let participants move on to completing the next questionnaire if there were questions left unanswered. Normality of data distribution was first assessed with the Shapiro–Wilk Test. Between group comparisons were undertaken with t tests for normally distributed data and Mann Whitney U tests for non-parametric data. Pearson and Spearman correlation coefficients were obtained for normally distributed and non-parametric data respectively.

2.2. Ethics approval

The study received approval from the Research and Ethics Board at Sunnybrook Health Sciences Centre fully affiliated with the University of Toronto. Project Identification Number: 5535.

3. Results

Demographic data: Fifty per cent of the sample was female. The mean age of the sample was 43.07 (SD = 11.83) years and they had been working on average 18.65 (SD = 11.17) years as journalists. Reporters (n = 106, 43.1%) made up the largest group of journalists. Further demographic data appear in Table 1.

Table 1.

Demographic characteristics of sample.

Variable	N = 246	Percentage	Mean	SD
Age	240		43.07	11.83
Sex
Male	118	48
Female	123	50
Non-binary/third gender	4	1.6
Gender (2SLGBTQI+)	34	13.8
Marital status
Single	52	21.1
In a relationship	64	26
Married	111	45.1
Separated	4	1.6
Divorced	11	4.5
Years in journalism or media			18.65	11.17
Type of journalism
Reporter	106	43.1
Editor	43	17.5
Manager or executive	18	7.3
Photographer	11	4.5
Video journalist	10	4.1
Other	58	23.6

Notes: The sample size for the demographic data is n = 246, ‘Prefer not to answer’ responses were excluded. The sample size for age is 240 due to 6 missing values.

Harassment data: The most widely used platform for online harassment was X (formerly Twitter; n = 155, 63%). Harassment at least weekly if not daily occurred in 65 (26%) of the journalists. The two main reasons for harassment given were the topic of the story covered, followed by accusations of fake news. The nature of the harassment included threats of physical harm, death, blackmail and rape, and warnings that family members would be targeted, amongst other threats. Gender was cited as of a reason for harassment by 44 (17.9%) journalists. Slightly more journalists viewed the online response to their work in a negative light (33.7% vs. 31.3%) whereas 22.4% thought the positives and negatives cancelled one another (n = 36, 14.6% of journalists did not answer this question). More detailed harassment data appear in Table 2.

Table 2.

Harassment data.

Variable	N = 246	Percentage	Mean	SD
Harassment platform
Twitter	155	63.0
Email	122	49.6
Facebook	93	37.8
Instagram	36	14.6
Text message/SMS	36	14.6
Frequency of harassment online
Never	43	17.5
Less than once every year	29	11.8
At least once a year	32	13.0
Once every 2 to 3 months	51	20.7
On a monthly basis	26	10.6
On a weekly basis	43	17.5
Every or almost every day	22	8.9
Reasons for harassment online
Stories covered	150	61.0
Media climate, including accusations of ‘fake news’	137	55.7
Media organization working for	118	48.0
The angle taken or way a story was reported	117	47.6
The changing reputation of journalists	96	39.0
Political position or ideology	74	30.1
My gender identity	44	17.9
Age	34	13.8
Ethnic or cultural background	32	13.0
Nationality	30	12.2
My sexual identity	18	7.3
Nature of harassment
Threats of physical harm	74	30.1
Vulgar/sexualized images/messages	52	21.1
Death threats	42	17.1
Family threatened	29	11.8
Personal image manipulated	25	10.2
Rape threats	18	7.3
Blackmail	18	7.3
Overall online responses to work are
More negative	83	33.7
More positive	77	31.3
Negative and positive responses cancel one another	55	22.4
Perceived level of support from news organization	101		5.80	3.01

Psychometric data: Although the mean anxiety (GAD-7) score of 7.54 (SD = 11.83) fell in the mild range, 74 (30.1%) of participants reported anxiety scores in the moderate to severe range (scores greater than 9). The mean depression (PHQ-9) score of 6.85 (SD = 5.83) fell in the mild range with 12.6% of the sample reporting depressive symptoms that were moderate to severe (scores from 15 to 19). One in six journalists (n = 39, 15.9%) reported PTSD symptoms above the threshold score for potential PTSD. A tentative diagnosis of PTSD according to symptoms scores of 2 or more from the four PCL-5 subscales was made in 40 (16.3%) journalists. A total of 34 (13.8%) journalists met both sets of criteria suggesting possible PTSD. More detailed psychometric data can be found in Table 3.

Table 3.

Psychometric data.

Variable	N = 246	Percentage	Mean	SD
GAD-7 Anxiety total scores			7.54	11.83
GAD-7 Anxiety severity scores
Minimal anxiety	94	38.2
Mild anxiety	78	31.7
Moderate anxiety	35	14.2
Severe anxiety	39	15.9
PHQ-9 total scores	246		6.85	5.83
PHQ-9 depression severity scores
None-minimal	113	45.9
Mild	68	27.6
Moderate	34	13.8
Moderately severe	19	7.7
Severe	12	4.9
PCL-5 symptom severity scores			15.76	16.34
DSM-5 symptom cluster severity scores
Intrusion			3.94	4.44
Avoidance			2.04	2.34
Negative Thoughts			5.27	6.20
Hyperarousal			4.51	5.17
PCL-5 cut-off scores
PTSD unlikely	207	84.1
Probable PTSD	39	15.9
TMIS-J total scores			10.93	7.07
Weekly units of alcohol			5.7	6.65
Past contact with a mental health specialist
Yes	165	67.1
No	78	31.7
Reason for contact with a mental health specialist
Both work-related and personal	78	47.6
Personal	56	34.1
Work-related	30	18.3
History of psychological treatment
Yes, for personal trauma	42	17.1
Yes, for both personal and work-related trauma	34	13.8
Yes, for journalism work-related trauma	21	8.5

The frequency of threats and harassment correlated significantly with anxiety (GAD-7; r = 0.16, p = .014), depression (PHQ-9; r = 0.15, p = .022), PTSD (PCL-5; r = 0.2, p = .002) and moral injury (TMIS-J; r = 0.3, p < .002).

The harassment and psychometric data were analysed according to sex. While there were no sex differences in frequency of harassment or believing harassment was becoming more frequent or vicious, women rated their lives in general and work lives in particular as more stressful than men. Women were significantly more anxious and depressed and had higher PTSD and moral injury scores than men (Table 4).

Table 4.

Sex differences and psychometric data.

Variable	Male (n = 118)		Female (n = 123)		t-test	Chi-square	p
	M	SD	M	SD
GAD-7 scores	6.06	5.16	8.78	6.14	3.51		<.001
PHQ-9 scores	5.47	4.93	7.94	6.23	3.12		.002
PCL-5 scores	12.72	14.24	18.48	17.75	2.89		<.004
TMIS-J scores	9.35	6.21	12.23	7.50	−3.24		<.001
Perceived increase in frequency of online harassment						.001	.99
Perceived increase in extremeness of online harassment						2.245	.134
Level of stress in work-life	6.60	1.97	7.24	1.75	−2.68		.008
Level of stress in life as a whole	5.81	1.90	6.50	2.05	−2.74		.007

Notes: M = mean. SD = standard deviation. GAD-7 = General Anxiety Disorder-7. PHQ-9 = Patient Health Questionnaire-9. PCL-5 = PTSD checklist for DSM-5. TMIS-J = Toronto Moral Injury Scale for Journalists. Level of stress in work-life and level of stress in life as a whole were rated on a scale from 0 (none) to 10 (a lot).

Thirty-four (13.8%) journalists identified as LGBTQ+. They reported statistically significantly more anxiety than their heterosexual colleagues (median = 7.0, range = 0–21 vs. median = 6.0, range = 0–21; Mann–Whitney U = 2723, p = .049) but not more depression, PTSD, or moral injury symptoms.

Moral injury (TMIS-J) correlated significantly with anxiety (GAD-7: r = .40), depression (PHQ-9: r = .41) and PTSD (PCL-5: r = .44; Spearman’s rho, p = .001 for all). Anxiety correlated significantly with depression (PHQ-9: r = .82) and PTSD (PCL-5: r = .51; Spearman’s rho, p = .001 for both). Depression (PHQ-9) correlated significantly with PTSD (PCL-5: r = .55, Spearman’s rho, p = .001). The rank order of emotions associated moral injury was anger (32%), shame and disgust (both 18.9%), guilt (17.1%), and contempt (13.1%).

Journalists who viewed the online response to their work in a negative light reported more depressive (F = 4.07; p = .018) and PTSD (F = 3.78; p = .025) symptoms.

Journalists rated the level of support they received from their news organization as modest (5.8/10). Journalists whose news organizations offered support in response to harassment (n = 139, 56.5%) had lower depression (PHQ-9: M = 6.42, SD = 5.24 vs. M = 8.41, SD = 7.36; t = −2.17, p = .035), PTSD (PCL-5: M = 14.14, SD = 15.03 vs M = 21.04, SD = 18.87; t = −2.29, p = .008) and moral injury (TMIS-J: M = 10.56, SD = 6.65 vs. M = 12.84, SD = 7.86; t = −2.05, p = .042) scores.

4. Discussion

Our data confirm that journalists are frequently harassed online. What differentiates our study from those preceding it is that we have elicited an array of psychological difficulties that are associated with the harassment. These include clinically significant levels of anxiety in approximately a third of participants, a potential diagnosis of PTSD in 14% of the sample, and clinically significant depression that involves one in eight journalists. Moreover, we have shown for the first time that online harassment is associated with quantifiable symptoms of moral injury. Notably, all these symptoms correlated significantly with the frequency with which journalists were harassed online.

The wide reach of online harassment of journalists is indicated by the geographical range of the studies reviewed: Portugal (Sampaio-Dias et al., 2023), the United Kingdom (Claesson, 2022), Nigeria (Uwalaka & Amadi, 2023), USA (Holton et al., 2023; Lewis et al., 2020; Miller, 2023; Miller & Lewis, 2022; Waisbord, 2020), Philippines (Tandoc et al., 2023), South Korea (Kim & Shin, 2025; Lee & Park, 2024), Australia (Jane, 2018), Latin America (Harlow et al., 2023), Germany (Menke & Seeger, 2024), Chile (Dodds et al., 2024), Sweden (Björkenfeldt, 2023; Löfgren Nilsson & Örnebring, 2016), India (Abhishek, 2022) and Finland (Celuch et al., 2023). A common thread that runs through all of these studies is a shared concern about the negative effects of harassment. To date these concerns have largely focused on defining the nature of the harassment (Miller, 2023), who is most at risk (Adams, 2018; Jane, 2018; Miller & Lewis, 2022; Tandoc et al., 2023), how harassment affects the way journalists work (Lewis et al., 2020) and the coping strategies used by to mitigate threats and the deleterious consequences of this (Harlow et al., 2023). The troubling picture that emerges may be broadly summarized as follows: social media violence directed against journalists can permeate all stages of news production; women and sexual minorities are particularly targeted; journalists self-censor, disconnect from aspects of their work, and contemplate leaving the profession; journalists’ reactions include anger, lethargy, and fear; coping strategies entail avoidance and emotion-focused approaches (generally regarded by therapists as maladaptive; Hori et al., 2024; Sigmon et al., 2007); and newsrooms in general are not doing enough to support journalists and address the challenges posed by online harassment.

Our study adds to these findings by revealing that a sample of Canadian journalists are harassed often with a severity that may meet the DSM–5 stressor criterion for PTSD (threats of death, rape, and physical violence; American Psychiatric Association, 2013), believe the harassment is becoming more frequent and vicious, and rate the support they receive from their news organization as modest, at best. This consistent body of negative findings helps frame the psychological findings reported here. Rates of anxiety and PTSD symptoms in particular are well above those documented in the general population (Patten et al., 2006; Van Ameringen et al., 2008). In keeping with a broader trauma literature, our data replicate the finding that anxiety, depression and PTSD symptoms are often comorbid (Sareen, 2014). This indicates that certain individual journalists are affected by multiple emotional difficulties. Furthermore, while the overall mean psychometric scores fall in the mild range, this obscures a sizeable minority of journalists with clinically significant symptoms of anxiety and PTSD in particular. Our finding of elevated anxiety replicates a study of Finnish journalists (Celuch et al., 2023) albeit more robustly for we utilized a valid psychometric scale to quantify this emotion. Similarly, elevated symptoms of PTSD found in our sample replicate Korean journalist data (Lee & Park, 2024).

A large segment of the journalism literature reveals the vulnerability of women (Adams, 2018; Claesson, 2022; Jane, 2018; Lee & Park, 2024; Sampaio-Dias et al., 2023; Tandoc et al., 2023), and particularly young women (Lewis et al., 2020), to harassment. Our data did not confirm this in keeping with a study from Nigeria (Uwalaka & Amadi, 2023), but we did find that women regarded their lives, including work lives, as significantly more stressful than men. We also showed that women were more anxious and depressed than men and had more symptoms of PTSD and moral injury. These observations in relation to anxiety, depression, and PTSD mirror the often replicated finding of similar sex differences in the general population (Christiansen & Berke, 2020; Farhane-Medina et al., 2022; Tolin & Foa, 2006).

A novel finding in our study pertained to moral injury. We were able to quantify this using the recently developed Toronto Moral Injury Scale for Journalists and our findings are again in keeping with the broader trauma literature that has shown an association between moral injury on the one hand, and PTSD, depression, and anxiety on the other. Interestingly, the most frequent emotion reported by journalists in response to moral injury was anger. A possible explanation for this is that the moral injury literature reveals that shame and guilt are more likely to arise when individuals behave in ways that transgress their own moral compass (Litz et al., 2009), whereas anger is more closely linked to the morally egregious behaviour of others (Zhang et al., 2024).

One potentially encouraging piece of data to emerge from our study is that two thirds of journalists had prior contact with a therapist. While this could in part reflect the relative ease of access to therapy in a Canadian medical system of universal health care coverage, it is also indicative of an openness on the part of journalists when it comes to managing their mental health. The downside of this observation however is that notwithstanding the provision of prior therapy, there remains a subgroup of journalists who continue to report elevated levels of psychopathology associated with harassment. Therapy, while necessary, is not the sole solution to the problem. Improved organizational support is needed, as our data and the journalism literature (Adams, 2018; Holton et al., 2023; Miller, 2023) suggests, together with implementing strategies other than avoidant and emotion-based coping that mitigate the more harmful aspects of harassment. These could include periodic education sessions on topics like moral injury and PTSD and easy, confidential access to a therapist. In relation to the latter, a notable finding from our study was that psychological support, when available, was associated with lower depression, PTSD and moral injury scores. This finding replicates data collected from news organizations during the pandemic (Osmann, Selva, et al., 2021) thereby providing additional, compelling reasons for news organizations to provide journalists with psychological assistance in response to work related traumatic stressors.

Our study is not without limitations. We do not know whether our sample is representative of the broader news organization surveyed. As such, the possibility of recruitment bias cannot be ruled out. In addition, our data collection relied solely on self-report measures. While the scores obtained in this fashion correlate strongly with psychiatric diagnoses arrived at through structured interview (Feinstein & Dolan, 1991), the two approaches are not synonymous. We were therefore unable to obtain formal psychiatric diagnoses in our sample. Finally, the wording of the PCL-5 stressor criterion does not replicate that in the DSM–5. It is therefore possible that the stressor reported by journalists as their most severe may not in all cases have passed the PTSD threshold required by the DSM–5.

In conclusion, our study is the first to show that online harassment is associated with a broad range of mental health difficulties including prominent symptoms of anxiety, depression and PTSD defined by the use of validated, widely used psychometric measurements. Adding to journalists’ emotional challenges is moral injury, not in itself a mental illness, but a condition allied to distressing emotions like anger, shame, guilt, disgust, and contempt. Our data align with the literature in recommending that news organizations should be doing more to address the challenge. The unique nature of our dataset adds another dimension to this recommendation by drawing attention to the adverse psychological consequences for journalists associated with online harassment.

Funding Statement

The website developed for the study was funded by the Canadian Broadcasting Corporation.

Disclosure statement

Anthony Feinstein reports grants from the MS Canada and the MacArthur and Knight Foundations, book royalties from Johns Hopkins University Press, Cambridge University Press, Amadeus Press and G Editions, and speaker’s honoraria from Novartis Merck and Roche. All other authors report there are no competing interests to declare.

Data availability statement

The data that support the findings of this study are available from the corresponding author, [AF], upon reasonable request.