Our research project “Hard to be a Hero” investigates the causes and consequences of perceived occupational heroism.

Some occupations are more likely than others to be perceived as being heroic. Firefighters, soldiers, police officers, or nurses, are often depicted in the media as heroic figures. Heroism, in itself, is defined through self-sacrifice: an heroic action is something that exposes one to risk, in order to help others. However, heroism is also defined as supererogatory actions: going above and beyond what is expected of you. Therefore, the notion of occupational heroes — workers heroised for doing their mission, become somewhat of a paradox (see Gölz, 2021).

Not much research has been done regarding heroic collectives. However, recent findings emphasise the relevance of heroism for exploitation of workers: heroised workers are perceived as more likely to accept exploitative policies (Stanley et al., 2023; Stanley & Kay, 2024). In one study, Stanley and Kay (2024, study 5) went one notch further by describing how manipulating occupational heroism resulted in increased support for exploiting the workers, in a sample of US respondents.

These findings urge us to understand the societal consequences of occupational heroism. As demonstrated during the Covid-19 pandemic, it is possible for occupations as a whole to be suddenly heroised in the media. Crises might be particularly fertile environments for campaigns heroising key-workers, and maybe to justify their sacrifice for the greater good when governments fail to respond in a manner that would protect all people.

Our research project therefore investigates both the causes and the consequences of perceptions of occupational heroism. It also questions the Janusian nature of occupational heroism: the ambiguous positive and negative outcomes associated with perceived heroism.

Five main outcomes associated with perceived occupational heroism were initially hypothesised:

H1: Perceived occupational heroism predicts higher gratitude toward workers

H2: Perceived occupational heroism predicts reduced criticism acceptability

H3: Perceived occupational heroism predicts reduced support for demands from the workers

H4: Perceived occupational heroism predicts reduced perception of victim-related aspects (e.g., suffering, vulnerability)

H5: Perceived occupational heroism predicts increased impunity and motivation to de-regulate the occupation

⚠️ Two of these hypotheses were later revised: H3: Perceived occupational heroism predicts increased support for demands from the workers

H4: Perceived occupational heroism predicts increased perception of victim-related aspects (e.g., suffering, vulnerability)

Complementing these revised hypotheses, three hypotheses were added:

H6: Perceived occupational heroism predicts increased perception that workers would choose to suffer rather than abandon the mission

H7: Perceived occupational heroism predicts increased perception that workers are willing to accept exploitative policies

However, these revisions were made after the development of the scale

Each of these five outcomes should be measured in a survey - as we are interested in public opinions. However, as far as moral judgments go, the contextualisation of the question can influence what participants report, with some authors claiming that contextualising moral judgments can increased the predictive validity of our measures (see Schein, 2020). On the other end, context-independent measures might be able to tap into beliefs that are applicable to a broader range of situations. We thus decided to define two levels of specificity of our measures for each target outcomes: a context-independent (general) and context-dependent (specific) measure. This approach further enable a robustness test of our hypotheses: if a variable predicts both outcomes, we can claim that the finding is robust to measurement type and report full support for our hypothesis – whereas a variable only predicting one of the two measures would mean only partial support for our hypothesis.

In order to develop a scale capturing these outcomes, we followed five main steps.

I. Initial items pool development

An initial pool of 82 items was generated in February 2025.

⚠️ Our scale is occupation-specific, that is: each scale can be applied to a specific target. To validate the scale, we focused on a ‘police officer’ version.

2. Construct validity assessment

A group of 5 experts in moral psychology and heroism study was consulted to provide feedback on the relevance of our items in measuring the target outcomes. Their feedback resulted in a 66-item questionnaire. This phase was completed in March 2025.

Data, material, analyses: https://osf.io/yzf69/files/osfstorage

3. Face validity assessment

A group of 20 naive UK residents evaluated the clarity of each item of the 66-items version of the scale. Their feedback led to a 54-item questionnaire. See Detailed Report on the Face validity assessment. This protocol was completed in April 2025.

Ethics: ID 202517441056499963 (Approved Apr 8, 2025)

Data, material, analyses: https://osf.io/yzf69/files/osfstorage

Full scientific report: https://jeanmoneger.com/uploads/facevalidityassessment_report

4. Reliability and Structure assessment

We asked 440 participants (representative UK sample) to complete the 54-item questionnaire. We then assessed the reliability (i.e., MacDonald’s omega) and the structure of the survey using Exploratory Factor Analyses. This Questionnaire assessment was registered. Our analyses led to a 21-item final scale that will be used for our research project. See Detailed Report on the Reliability and Structure assessment. This study was completed in May 2025.

Ethics: ID 202517473026909997 (Approved May, 15, 2025)

Registration: https://osf.io/cs23a/overview

Data, material, analyses: https://osf.io/yzf69/files/osfstorage

Full scientific report: https://jeanmoneger.com/uploads/registered_factor_analyses

5. Definitions of occupations specific dilemmas

In our scale, the measure of specific support for impunity uses target specific dilemmas.

For each target occupation, we had to design a specific dilemma situation where regulations are at odds with the completion of the worker’s mission. Each dilemma started with a paragraph describing regulations applying to the occupation, and then a paragraph describing a worker who decided to ignore the regulations in order to do their work. For instance, participants assigned to the journalist condition could read:

UK journalists must follow strict codes of ethics, such as IPSO rules and their employers’ policies. These ban phone hacking or wiretapping except in rare, extreme cases. The rules stress respect for privacy, following the law, and avoiding unjustified intrusion, even for the public interest.

A national journalist suspects a top official of abusing their position by steering public contracts to companies owned by close associates, raising concerns of corruption and misuse of public money. With no evidence and official investigations blocked, the journalist believes normal reporting won’t work fast enough. They secretly tap the phones of the official, his wife, and two daughters for several weeks, hoping to find proof.

In order to extend our scale to other occupations of interest (i.e., soldiers, police officers, nurses, firefighters, journalists, psychiatrists, and underwater welders), we had to design specific dilemmas.

We surveyed 300 participants (representative UK residents) and asked them to evaluate the heroism of soldiers, nurses, firefighters, journalists, psychiatrists and nurses (the police officer version being evaluated in step 4 already). They then filled the general support for impunity for each of these targets, as well as the specific dilemmas developed for each of the target occupations. In this study, we observed that correlations between the general and the specific measures within each target occupations varied between r = .32 (in the Psychiatrist condition) and .42 (in the Firefighter condition).

Pairwise Wald tests (with Holm adjustment) between each correlations for each occupations indicated that no correlation significantly differed from another — thus supporting invariance between the specific dilemmas and the general support for impunity measures across all target occupations.

Ethics: 2025175682910010087 (Approved September, 02, 2025)

Registration: https://osf.io/2xtn8/overview

Data, material, analyses: https://osf.io/bjpgd/overview

Full scientific report: https://jeanmoneger.com/uploads/hero_moralshielding_sept01


Full scale description

Target group description

Each section of the questionnaire contains a header describing the target occupation in a generic way, using one sentence. For instance:

Journalists
Journalists investigate and report on stories to inform the public.

All items referred to this target group.

Response scales

Several seven-point response formats were used across the survey.

Response format Anchors
Likelihood 1 = Very unlikely, 2 = Quite unlikely, 3 = Slightly unlikely, 4 = Neither likely, nor unlikely, 5 = Slightly likely, 6 = Quite likely, 7 = Very likely
Agreement 1 = Strongly disagree, 2 = Moderately disagree, 3 = Slightly disagree, 4 = Neither disagree, nor agree, 5 = Slightly agree, 6 = Moderately agree, 7 = Strongly agree
Extent 1 = Not at all, 2 = Very little, 3 = A little, 4 = Somewhat, 5 = Quite a bit, 6 = A lot, 7 = Very much
Frequency / justification 1 = Never, 2 = Very rarely, 3 = Rarely, 4 = Occasionally, 5 = Frequently, 6 = Very frequently, 7 = Always

Regarding the Specific Criticism acceptability, Checklist items were presented as binary selections, where participants could select any number of options, including none.

Overview of measures

Construct Definition Format Number of items
Specific gratitude Material/behavioral expressions of gratefulness (donating, volunteering, public displaying) Likelihood scale 3
General gratitude General gratefulness for their work Extent scale 1
General Criticism acceptability Opinion towards critics Agreement scale 3
Specific Criticism acceptability Reactions to online hate speech Checklist responses to vignettes 2 vignettes × 5 options
Specific Support for workers’ demands Support for protesting action Agreement scale 2 items + 1 attention check
General Support for workers’ demands Support for Worker vs Government initiated propositions for improvement of the occupation Frequency / justification scale 2
General Victimisation Perceived victimization (exploited, unfairly treated, victimized) Extent scale 3
Specific Victimisation Need for protection, decreased expectations, and perceived resilience to pain Agreement scale 3
General support for impunity Support for de-regulating the occupation Agreement scale 3 items + 1 attention check
Specific support for impunity Support for protecting workers from prosecutions Agreement scale 3 items + 1 attention check

Attention-check items were included for data-quality screening and should not be scored as part of the substantive scales.

Full item wording

Specific gratitude

Participants were asked:

If there were a public campaign in support of [Target Occupation], how likely would you be to do each of these things in response?

Responses were given on the seven-point likelihood scale.

Suggested item label Item wording
symbolic_gratitude_1 Sharing a supportive post about [Target Occupation] on my social media
symbolic_gratitude_2 Donating to a campaign that aims to raise public awareness about the contributions of [Target Occupation]
symbolic_gratitude_3 Volunteering time to help with initiatives that support the work of [Target Occupation]

General gratitude

Participants were asked:

To what extent do you feel grateful for [Target Occupation]’ work?

Responses were given on the seven-point extent scale.

Suggested item label Item wording
single_item_gratitude To what extent do you feel grateful for [Target Occupation]’ work?

General criticism acceptability

Participants were asked:

To what extent do you agree with the following statements?

Responses were given on the seven-point agreement scale.

Suggested item label Item wording
general_critic_1 People should think twice before they criticize [Target Occupation]
general_critic_2 People should stop badmouthing [Target Occupation]
general_critic_3 Negative comments about [Target Occupation] often go too far

Specific criticism acceptability

Participants were asked:

For each social media post below, please read the statements underneath it, and tick all that you agree with. You may select as many or as few as you like, including none.

The following two social media posts were presented.

Vignette label Post wording
specific_critic_post_1 An online post that says: “[Target Occupation] should die and burn in hell”
specific_critic_post_2 An online post that says: “[Target Occupation] are evil and wish harm on other people”

For each post, participants could select any of several possible response options. Responses are coded for severity, and for each participant, we score the item as the maximal severity.

Response option Severity
This post should be liked 1
This post should be ignored, but they have a right to say that 2
This post should be deleted from the social media platform by moderators 3
The person making the post should be banned from the platform by moderators 4
The person making this post should be prosecuted using the UK laws against “grossly offensive” public messaging 5

Specific Support for workers’ demands

Participants were asked:

To what extent do you agree with the following statements?

Responses were given on the seven-point agreement scale.

Suggested item label Item wording Item type
support_demand_1 [Target Occupation] should protest more for the rights they deserve Substantive
support_demand_attention To show you’re paying attention, please select ‘Moderately disagree’ for this statement. Attention check
support_demand_2 I support [Target Occupation] who choose protest as a way to achieve fair working conditions Substantive

General victimisation perception

Participants were asked:

How much do you see [Target Occupation] as:

Responses were given on the seven-point extent scale.

Suggested item label Item wording
victim_perception_1 Victimised
victim_perception_2 Unfairly treated
victim_perception_3 Exploited

Specific victimisation perception

Participants read the following prompt:

Consider the following observation from a recent report: In their professional life, more than 60% of [Target Occupation] have reported intense migraines from working long hours. How much would you agree or disagree with the following statements:

Responses were given on the seven-point agreement scale.

Suggested item label Item wording
harm_report_1 I believe [Target Occupation] are strong enough to face this condition
harm_report_2 It shows that we are asking too much of [Target Occupation]
harm_report_3 This report shows that [Target Occupation] require greater protection

General support for impunity

Participants were asked:

In your own personal opinion, how much would you agree or disagree with the following statements:

Responses were given on the seven-point agreement scale.

Suggested item label Item wording Item type
permissiveness_1 [Target Occupation] should be given more freedom in the way they do their work Substantive
permissiveness_2 [Target Occupation] can do their job better when they’re not forced to follow standard procedures Substantive
permissiveness_attention Please tick “Moderately disagree” to demonstrate that you are paying attention to the survey Attention check
permissiveness_3 [Target Occupation] should be able to do whatever it takes to achieve their mission Substantive

Specific support for impunity

For each target occupation, we had to design a specific dilemma situation where regulations are at odds with the completion of the worker’s mission. Each dilemma started with a paragraph describing regulations applying to the occupation, and then a paragraph describing a worker who decided to ignore the regulations in order to do their work. For instance, participants assigned to the journalist condition could read:

UK journalists must follow strict codes of ethics, such as IPSO rules and their employers’ policies. These ban phone hacking or wiretapping except in rare, extreme cases. The rules stress respect for privacy, following the law, and avoiding unjustified intrusion, even for the public interest.

A national journalist suspects a top official of abusing their position by steering public contracts to companies owned by close associates, raising concerns of corruption and misuse of public money. With no evidence and official investigations blocked, the journalist believes normal reporting won’t work fast enough. They secretly tap the phones of the official, his wife, and two daughters for several weeks, hoping to find proof.

Participants were then asked:

Considering this situation, how much do you agree with the following statements:

Responses were given on the seven-point agreement scale.

Suggested item label Item wording Item type
dilemma_attention To show you’re paying attention, please select ‘Slightly agree’ for this statement. Attention check
dilemma_protection_1 This [Target Occupation] should be protected from prosecution for this violation Substantive
dilemma_protection_2 There should be a law shielding this [Target Occupation] from civil lawsuits in this situation Substantive
dilemma_protection_3 This [Target Occupation] should be protected from disciplinary sanctions imposed by their professional body for this violation Substantive

General support for workers’ demands

Participants read the following prompt:

Publicly funded jobs, like those of [Target Occupation], can be improved in many ways — such as raising salaries, hiring more staff, or upgrading facilities. When resources are scarce, these needs are not always met. Sometimes, the government takes the initiative, recognises a need for improvement, and invests time and resources into making it happen. Other times, journalists take the initiative and pressure the government for improvements they think they need. For some proposals, it might feel more appropriate for the government to take the lead; for others, it may seem more fitting for the workers to do so.

Participants were then asked:

In the UK, how often do you think it is justified for…

Responses were given on the seven-point frequency / justification scale.

Suggested item label Item wording
government_lead_improvements The government, and not [Target Occupation], to take the lead on pushing improvements in the profession?
journalists_lead_improvements The [Target Occupation], and not the government, to take the lead on pushing improvements in the profession?