Research on Unrealistic Optimism among HoReCa Workers as a Possible Future Hotspot of Infections

: As we are facing a new surge of the highly infectious delta variant of COVID-19, there is an urgent need for research to reduce the harm before this next wave hits. In the present paper, we present data that is alarming. We have found that HoReCa (hotels, restaurants, and catering services) workers, who are highly exposed to many new social interactions in close contact, present an unrealistic optimism (UO) bias: they perceive themselves as less at risk to this virus in comparison to others. From the literature, we already know that individuals holding this view are less involved in preventive actions and present more risky behaviors. In the face of the delta variant, this leads to the conclusion that restaurants will be new hot spots. What is more, we found that these unrealistic expectations are more pervasive: workers of the restaurant industry estimate low chances of bankruptcy, which may lead to unrealistic salary expectations, leading owners to a new upcoming wave of crisis: COVID-19 and bankruptcy—both of which may be caused by their workers.


Introduction
The COVID-19 outbreak has deeply affected the world's economy in general [1] and the hospitality industry, in particular [2]. Our routines have been deeply disrupted, some of us have experienced intense anxiety that destroys our day-to-day plans, while others suffer from isolation, uncertainty, boredom, and serious health concerns. Additionally, some people have been unable to get to stores, while others need to purchase regular medicine but may not be able to order prescriptions by phone or online. It has also been harder for many of us to do the things that normally help keep us well and to make plans about shopping and vacation/business trips. From this perspective, the hospitality industry was hit incredibly hard; businesses have been or even will be closed for months; revenue is lost, and customers may not come back for a long time. Such closures are for a good reason. Thus, although it has been shown that mental health and coping with stress during the COVID-19 pandemic depends on the individual's specific type of disease rather than their type of profession [3], in the present paper, we reverse this thinking by concentrating on a specific profession: restaurant workers. The rationale for this perspective is that the aforementioned seminal papers bring an individual perspective, whereas this study focuses on groups as a potentially important factor in the COVID-19 pandemic. Taken together, both works may fill an important gap in research on COVID-19 from the perspective of a sustainable work and business environment.
For example, a large number of new infections took place on cruise ships [4], in ski resorts, and in restaurants and bars [5]. As a result, business models changed dramatically to take-out only, or restaurants reopened with a limited capacity [6].
One should keep in mind that there is also another side to the story: workers becoming front-line soldiers to make the aforementioned services available. Someone has to work in healthcare businesses, and it is often easy to think of those as front-line workers. There is, however, another crucial front-line worker that seems to be overlooked: workers in the hospitality industry, without whom travelers could not stay in a hotel, eat in a restaurant, or have a drink in a pub. The goal of this paper is to bring this group into focus, especially those who work in the HoReCa industry: wait staff, bartenders, caterers, kitchen staff, etc. These essential workers are especially vulnerable to COVID-19 infection since guest turnover is high. On the one hand, these people need to work for a living; on the other hand, they need to stay alive and provide security for their families. In this paper, we present the possible cognitive bias that makes work in such an industry possible, but still potentially extremely dangerous. This perspective is especially important while we are facing the next worldwide surge: the delta variant.
With a huge number of confirmed COVID-19 cases and deaths affecting the world's population, governments worldwide have implemented strict social distancing regulations to help reduce the spread of the virus. These regulations included steps to reduce social contact in hotels, bars, restaurants, pubs, cafeterias, bakeries, etc. Additionally, all citizens had to cancel non-essential travel, some of which will not be possible for many months ahead. It is obvious that the coronavirus affected, affects, and will affect all levels of the hospitality industry. It is also apparent, however, that not everyone has the same capacity for recovery. Undoubtedly, some hotels, restaurants, or pubs that have closed may never reopen. To make matters worse, survivors will most likely employ fewer staff. This is why, while all sectors of the global economy may be affected by the COVID-19 outbreak, the hospitality industry has been hit particularly hard, and a plan for recovery may not be effective or clear.
People working in the hospitality industry have an additional reason-personal COVID-19 infection-to feel discomfort during a pandemic. They may worry about the fate of the companies they work for and about their own professional futures, as well as the financial security of themselves and their families. In this paper, we present a bias making those future predictions more optimistic, but, by the same token, leading people to hold these views may lead to more exposure to COVID-19 infection, and most importantly, this bias may make these people less eager to continue social distancing, washing hands, and wearing face masks. This negative tendency may result in creating hotspots of future infections.

Unrealistic Optimism/Pessimism Bias and Its Presence during the COVID-19 Pandemic
People are sometimes called unrealistically optimistic when they predict that future events will be more favorable than suggested by a quantitatively objective standard. For example, we can talk about unrealistic optimism when someone expects a high grade in an exam but actually receives a low grade due to poor preparation [7]. Similarly, we can-post factum say that economists were unrealistically optimistic about their predictions concerning inflation [8], and most people are unrealistically optimistic when it comes to estimating the time they need to perform various activities [9].
A specific form of unrealistic optimism falls under the empirical scrutiny of psychology, namely when people predict that their personal outcomes will be more favorable than the outcomes of their peers. Th roots of unrealistic expectations are clear: it is obvious that at the group level, people who report below-average risk should be balanced by the same number of people reporting above average risk. If, however, the vast majority of people in any group consider themselves less vulnerable to future negative events than the average person in that group, then a large number of these people simply cannot be right [10]. This is why this bias was coined unrealistic optimism.
It is clear then, in the psychology domain, contrary to other scientific fields mentioned above, that we are dealing with a comparative mechanism. One should take into account, however, that in this case, the adjective "unrealistic" refers to a group, not to a specific individual. For example, someone who says that he or she is less likely to get lung cancer than the average person might be right if no one in his or her family has had cancer and this person has not smoked her/his entire life. In our study, we decided to investigate precisely such a form of unrealistic optimism as a future predictor in the development of the COVID-19 pandemic.
In detail, the psychological literature describes this phenomenon when people estimate the probability of different life events as the belief "that negative events are less likely to happen to them than to others, and they believe that positive events are more likely to happen to them than to others" [11]. It has been demonstrated, for example, that people typically rate themselves as less likely than other people to fall victim to heart disease [12], divorce [13], alcohol dependence [14], or be involved in a car accident [15]. A recent study conducted by Ngan and Tze-Ngai Vong [16] revealed that hospitality employees at the entry-level position are optimistically biased regarding their promotion expectations. Lopez and Leffingwell [17], in turn, presented evidence that college students hold a whole set of positively biased beliefs concerning risky sexual behavior. Participants believed they had a significantly lower personal risk of engagement in alcohol-related sexual activity, unintended pregnancy, and sexually transmitted infections than the average college student. One may conclude that there is robust evidence that normal, mentally healthy people's perceptions of reality are characterized not by an accurate outlook on the future, but by biased and self-serving judgments.
This common bias, called unrealistic optimism, is defined in a seminal article written by Taylor and Brown [18] as a part of a pattern of so-called positive illusions that help us cope with potentially threatening situations. Indeed, there is broad empirical evidence that unrealistic optimism reduces stress, fear, and anxiety [19,20], and helps the person to retain a sense of personal control [21][22][23].
In two recently published studies, unrealistic optimism has also been found to be present in the COVID-19 era. Firstly, it was demonstrated that study participants held an unrealistic optimism view not only before the first cases were reported in Poland (the country of the study), which sounds reasonable; one may expect not to be affected by the pandemic since it is not present, and after, when the first cases were reported in this area. Interestingly, unrealistic optimism bias was also reported, and was constant during the massive spike in cases [24]. Secondly, the presence of unrealistic optimism was reported in different cultures and countries [25], making the possible critique of the uniqueness of the pattern of the result unfeasible.
However, when people are directly confronted with highly accessible and very salient information that contradicts these positive illusions, a reverse bias may occur. Dolinski and colleagues [26] examined reactions among Polish citizens after their exposure to radiation following the Chernobyl nuclear plant disaster in 1986. Participants believed they were more (not less) likely than the average person to suffer health problems as a result of intense radiation. Authors of this paper dubbed this bias unrealistic pessimism. A similar pattern of results was obtained by [27]. People who experienced the 1989 California earthquake were asked a few days afterward to assess the likelihood that they and the average person their age would someday be injured in a natural disaster, such as an earthquake; the participants demonstrated no evidence of unrealistic optimism.
Significantly, ref. [26] study demonstrates that participants holding an unrealistic pessimism bias presented very active behaviors, preventing harm to their own health in every possible way to make the radiation less lethal. They, more often than optimists, purchased and drank Lugola liquid, did not leave their homes, secured their windows, and abstained from buying milk and vegetables, which were especially susceptible to becoming polluted from the radioactive cloud that was becoming a deadly poison.

The Hospitality Industry in a COVID-19 Era
The hospitality industry plays a crucial role in the development of the US economy, supplying 20% of the GDP in 2017, and providing 30% of workplaces [28]; this sector of the economy was hit incredibly hard, as demonstrated by the 53% drop in employment between the middle of February and middle of April 2020 [29]. Analysis provided by the McKinsey Company reveals that in the US alone, 44 to 57 million jobs will be affected (and mostly in the hospitality industry: food and accommodation services) by the COVID-19 pandemic [30]. A prediction reported by Huang and colleagues [31] estimates a 20-30% reduction in food/drink and leisure/entertainment sectors in one month alone at the beginning of the COVID-19 pandemic. As the review by Gursoy and Chi [6] clearly demonstrates, not only is the hospitality industry in a deep crisis as a repercussion of COVID-19, but it will also face significant challenges as a result of this global crisis, in terms of the economy, politics, and global health issues. It seems that clients of restaurants (64.71%) expect delivery of new services in customer waiter/waitress relationships.
HoReCa workers wear facemasks, implement strict regulations like social distancing and obligatory hand sanitizing, reduce the number of clients, and disinfect high-touch surfaces. Training staff in health protocols is simply not enough; expected new solutions include menus delivered in a way that eliminates touch (digital menus) and touchless payments, suggesting that new solutions have to be provided to upgrade accessibility to those services. What is more, the majority of study participants declared that until such changes are provided, they will not dine in a restaurant [32]. A call for new solutions to be found stemming from behavioral research in hospitality was expressed in a review by Gursoy and Chi [6] and the below reported study tries to address this vital call. What is more, the study reported here addresses two other important issues: (1) it may enable better management with the delta variant surging soon, and (2) it may help HoReCa businesses avoid massive closures not only due to lockdowns, but also from the erroneous financial expectations of workers, leading to possible bankruptcies.

The Goal of the Study
One can say, therefore, that in situations where there is only a potential for danger, and it concerns a distant future, unrealistic optimism is beneficial: it helps in dealing with highly dangerous situations, thus making them less threatening. If it is real and current, however, this bias will be harmful [33][34][35][36]. In relation to people employed in the hospitality industry, it would be beneficial to display unrealistic optimism about their professional future and financial security, but not to display it about the possibility of contracting coronavirus. In our study, we decided to investigate the beliefs of hospitality industry employees regarding the chances of various negative states of affairs occurring during and after a pandemic.
The goal of the present paper is also to assess the bias held by hospitality workers, and their (in comparison to others') expectations about health risks stemming from the hospitality industry (i.e., direct exposure through a high degree of social contact), and the security of their workplaces and industry. This is especially important due to the fact that, as mentioned above, unrealistic optimism might, one the one hand, lead to more eagerness to come back to work, but on the other hand, it might affect those people's health by being too optimistic and result in exposing one's own health.
Obviously, assessing one's own chances of avoiding different negative states of affairs may vary from case to case; so, a positive outcome would be if employees of the hospitality industry demonstrated only partial invulnerability. Precisely, it would be beneficial for them if they displayed unrealistic optimism regarding the possibility of keeping their jobs, industry development, and survival of the companies for which they work, but at the same time without displaying unrealistic optimism regarding coronavirus infection (which would cause them to be more vulnerable to infection and its health-related repercussions).
From a broader perspective, however, unrealistic optimism concerning the economic situation might also be harmful. This bias may influence workers' salary expectations; they may demand increased spending on their wages. As such expectations cannot be met during a crisis, this may result in lower employee satisfaction. As a result, HoReCa workers may also be less loyal to their employers and quit their jobs [16,37]. Finally, assessing psychological factors prior to the next surge stemming from the delta variant may be crucial for assessing new directions for management predictions.
The body of literature on unrealistic optimism in COVID-19 presented above clearly indicates that COVID-19 is a robust phenomenon. However, even though it was found to exist in many areas of social life (e.g., students [24]), research on medical healthcare providers demonstrates that they do not present unrealistic optimism bias [25]. On that basis, this study aimed to assess this bias among HoReCa workers as a future prediction of danger (in the case of presence of the bias), or just the opposite: a signal of safety.
Taken together, in this research we expected that HoReCa workers would exhibit unrealistic optimism when comparing themselves with other workers of this specific industry (to be illusional). At the same time, we expected that HoReCa workers would not exhibit unrealistic optimism bias regarding the general public and fellow countrymen since the specificity of this business brings great risk (many interactions; a business model widely affected by the COVID-19 pandemic; to be delusional).
In light of the abovementioned arguments, the following hypotheses were formulated: Hypothesis #1: A HoReCa worker, when comparing her/himself to other workers from that specific industry, will exhibit unrealistic optimism (both in terms of health and employment). Hypothesis #2: A HoReCa worker will not exhibit unrealistic optimism (again: both in terms of health and employment) when comparing her/himself to the general public. This is due to the specificity of the industry.
Other predictions (concerning, for example, the eagerness to vaccinate) were left as open questions without posing hypotheses and/or predicting a precise direction of the relationship.

Method Participants and Design
A sample of Polish gastronomy workers was recruited for an online study. Owners of the restaurants were asked to pass the link to the study to their workers. In the next step, to rule out the possibility that the pattern of the results is specific only to one location, we asked the participants to pass the link to workers from different restaurants.
To estimate the sample size a priori, we calculated the total sample size based on an effect size of 0.25 and p level = 0.05. To achieve a power of 0.95 for this effect size, the estimated sample size was N = 175 [38]. Since there was a possibility of attrition, we ran the study with a slightly larger sample to address the estimated sample size.
One-hundred eighty-one participants (93 women, 88 men, M age = 31.72, SD age = 11.25), whose age ranged from 20 to 71, agreed to take the survey. Thirty-one participants (16 women, 15 men, M age = 31.81, SD age = 10.96), whose age ranged from 21 to 63, were excluded from the analyses due to a lack of answers (which we treated as a proxy of lack of attention). Thus, the final sample consisted of 150 participants (77 women, 73 men, M age = 31.7, SD age = 11.34), who were aged 20 to 71. The gender distribution is equidistant: χ 2 (150) = 0.11; p = 0.744. The data were collected from 03 June 2020 to 15 June 2020.

Materials
Participants rated their agreement (1 = very low; 11 = very high) with the following twelve statements displayed in four separate blocks (depending on the main topic for each block), consisting of three questions. After answering the last question in the last block, respondents were thanked, asked about their gender and age, and finally, the survey ended. Block 2-assessment/comparisons for self-protection by vaccination. (Q4) How likely is it that you will get vaccinated for coronavirus once a vaccine becomes available on the market? (Q5) How likely is it that another person in your business gets vaccinated for coronavirus once a vaccine becomes available on the market? (Q6) How likely is it that a fellow countryman in your country of residence gets vaccinated for coronavirus once a vaccine becomes available on the market? Block 3-assessment/comparisons for losing a job. (Q7) What are your chances of losing a job because of the coronavirus pandemic? (Q8) What are the chances that another person in your business will lose a job because of the coronavirus pandemic? (Q9) What are the chances that a fellow countryman in your country of residence will lose a job because of the coronavirus pandemic?
Block 4-assessment/comparisons for bankruptcy in the industry. (Q10) What are the chances that the company you are working for will go bankrupt because of the coronavirus pandemic? (Q11) What are the chances that the average company in your business will go bankrupt because of the coronavirus pandemic? (Q12) What are the chances that the average company in your country of residence will go bankrupt because of the coronavirus pandemic?

Unrealistic Optimism/Pessimism Assessment for COVID-19 Infection
To test hypothesis #1, a within-subjects ANOVA was run. The results displayed statistically significant differences: F(2, 148) = 7.3; p = 0.001, η p 2 = 0.09. The analysis of multiple comparisons with the Bonferroni correction demonstrated that HoReCa workers believe that it is more likely that another person in their business will catch the virus (M = 4.99, SD = 2.71), than themselves (M = 4.53, SD = 2.77; p = 0.005). This result confirms hypothesis #1. A similar relationship was observed when compared to a fellow countryman, which did not support hypothesis #2. Again, gastronomy workers found it more likely that their fellow countrymen would be more likely to catch the virus (M = 5.23, SD = 2.8), not them (p = 0.003), meaning, in both cases, that unrealistic optimism was at play. Interestingly, no statistically significant differences were observed when assessing the likelihood of a colleague and a fellow countryman catching the virus (p = 0.591) (Figure 1).

Assessment/Comparisons for Self-Protection by Vaccination
A within-subjects ANOVA displayed statistically significant differences: F(2, 148) = 7.66; p = 0.001, η p 2 = 0.09. The analysis of multiple comparisons with the Bonferroni correction displayed no statistically significant differences when estimating the likelihood of a vaccine being taken between gastronomy workers (M = 5.61, SD = 3.54) and another person in their business (M = 5.23, SD = 2.73; p = 0.239). Additionally, gastronomy workers do not believe that a fellow countryman was more likely to get vaccinated (M = 6.05, SD = 2.46) than them (p = 0.414). However, this analysis displayed one statistically significant difference. It turns out that gastronomy workers believe that it is more likely that a fellow countryman will get vaccinated than another person in their business (p = 0.001) ( Figure 2); this means that in the case of vaccination tendencies, no bias was found (for neither unrealistic optimism nor pessimism). Note. Bars represent mean values; error bars represent standard error of mean; ** p < 0.01.

Assessment/Comparisons for Self-Protection by Vaccination
A within-subjects ANOVA displayed statistically significant differences: F(2, 148) = 7.66; p = 0.001, ηp 2 = 0.09. The analysis of multiple comparisons with the Bonferroni correction displayed no statistically significant differences when estimating the likelihood of a vaccine being taken between gastronomy workers (M = 5.61, SD = 3.54) and another person in their business (M = 5.23, SD = 2.73; p = 0.239). Additionally, gastronomy workers do not believe that a fellow countryman was more likely to get vaccinated (M = 6.05, SD = 2.46) than them (p = 0.414). However, this analysis displayed one statistically significant difference. It turns out that gastronomy workers believe that it is more likely that a fellow countryman will get vaccinated than another person in their business (p = 0.001) ( Figure  2); this means that in the case of vaccination tendencies, no bias was found (for neither unrealistic optimism nor pessimism).

Assessment/Comparisons for Losing a Job
A within-subjects ANOVA displayed statistically significant differences: F(2, 148) = 27.17; p < 0.001, η p 2 = 0.27. The analysis of multiple comparisons with the Bonferroni correction demonstrated that gastronomy workers believe that it is least likely that they will lose their jobs (M = 4.77, SD = 3.01), followed by another person in their business (which supports hypothesis #1; M = 5.93, SD = 2.67), and that their fellow countrymen are most likely to lose their jobs (which did not support hypothesis # 2; M = 6.49, SD = 2.22). Statistically significant differences were observed between all pairs: myself vs. another person in their business (p <0.001), myself vs. fellow countryman (p < 0.001), and another person in their business vs. fellow countryman (p = 0.026) (Figure 3). rection demonstrated that gastronomy workers believe that it is least likely that they will lose their jobs (M = 4.77, SD = 3.01), followed by another person in their business (which supports hypothesis #1; M = 5.93, SD = 2.67), and that their fellow countrymen are most likely to lose their jobs (which did not support hypothesis # 2; M = 6.49, SD = 2.22). Statistically significant differences were observed between all pairs: myself vs. another person in their business (p <.001), myself vs. fellow countryman (p < 0.001), and another person in their business vs. fellow countryman (p = 0.026) (Figure 3).

Assessment/Comparisons for Bankruptcy in the Industry
A within-subjects ANOVA displayed statistically significant differences: F(2, 148) = 43.71; p < 0.001, ηp 2 = 0.37. The analysis of multiple comparisons with the Bonferroni correction demonstrated that gastronomy employees believe that it is less likely that their company will go bankrupt (which supports hypothesis #1; M = 4.17, SD = 2.58) than the average company in their industry (M = 6.07, SD = 2.24; p < 0.001). A similar relationship was observed when compared to the average company in their country which did not support hypothesis #2. Again, gastronomy workers find it more likely that the average company in their country will go bankrupt (M = 5.71, SD = 2.03), not their own company (p < 0.001). Surprisingly and unexpectedly, no statistically significant difference was observed in assessing the likelihood of bankruptcy for an average company in their country and an average company in their industry (p = 0.190) (Figure 4). This means that, contrary to our expectations expressed in hypothesis #2, HoReCa workers perceived themselves as less exposed, in comparison to fellow countrymen, to danger expressed by the risk of losing a job.

Assessment/Comparisons for Bankruptcy in the Industry
A within-subjects ANOVA displayed statistically significant differences: F(2, 148) = 43.71; p < 0.001, η p 2 = 0.37. The analysis of multiple comparisons with the Bonferroni correction demonstrated that gastronomy employees believe that it is less likely that their company will go bankrupt (which supports hypothesis #1; M = 4.17, SD = 2.58) than the average company in their industry (M = 6.07, SD = 2.24; p < 0.001). A similar relationship was observed when compared to the average company in their country which did not support hypothesis #2. Again, gastronomy workers find it more likely that the average company in their country will go bankrupt (M = 5.71, SD = 2.03), not their own company (p < 0.001). Surprisingly and unexpectedly, no statistically significant difference was observed in assessing the likelihood of bankruptcy for an average company in their country and an average company in their industry (p = 0.190) (Figure 4). This means that, contrary to our expectations expressed in hypothesis #2, HoReCa workers perceived themselves as less exposed, in comparison to fellow countrymen, to danger expressed by the risk of losing a job.

Correlation Analysis
We also calculated the correlations between the specific questions for self-estimations to measure how interlinked these dimensions are: (Q1) How do you rate/judge the risk of being infected by coronavirus? (Q4) How likely is it that you will get vaccinated for coronavirus once a vaccine becomes available on the market? (Q7) What are your chances of losing a job because of the coronavirus pandemic? (Q10) What are the chances that the company you are working for will go bankrupt because of the coronavirus pandemic?

Correlation Analysis
We also calculated the correlations between the specific questions for self-estimations to measure how interlinked these dimensions are: (Q1) How do you rate/judge the risk of being infected by coronavirus? (Q4) How likely is it that you will get vaccinated for coronavirus once a vaccine becomes available on the market? (Q7) What are your chances of losing a job because of the coronavirus pandemic? (Q10) What are the chances that the company you are working for will go bankrupt because of the coronavirus pandemic?
We obtained two significant mediocre and positive correlations (0.3 < r < 0.6) between a desire to be vaccinated for coronavirus (Q4) and the estimation of the risk of being infected (Q1), as well as between the estimation of the chance that their company will go bankrupt (Q10) and the estimation of the probability of losing a job (Q7). We obtained one more significant but weak correlation (r = 0.19) between the estimation of the chance that their company will go bankrupt and the estimation of the risk of being infected (see Table 1). In the next step, we calculated the correlations between those specific items dealing with links between estimations about co-workers: (Q2) How do you rate/judge the risk of another person in your business being infected by coronavirus? (Q5) How likely is it that another person in your business gets vaccinated for coronavirus once a vaccine becomes available on the market? (Q8) What are the chances that another person in your business will lose a job because of the coronavirus pandemic? (Q11) What are the chances that the average company in your industry will go bankrupt because of the coronavirus pandemic?
Again, we obtained two significant mediocre and positive correlations (0.3 < r < 0.4) between the estimation of the chance that another person in their business will be vaccinated for coronavirus (Q5) and the estimation of the risk of another person in their business being infected (Q2), as well as between the estimation of the chance that another company will go bankrupt (Q11) and the estimation of the probability that another person will lose a job (Q8).
We obtained one more significant but weak correlation (r = 0.23) between the estimation of the chance that their company will go bankrupt (Q11) and the estimation of the risk of being infected (Q2). This time, another correlation between the estimation of the risk of another person in their business being infected (Q2) and the estimation of the probability that another person will lose a job (Q8) turned out to be significant but weak (r = 0.2; see Table 2). In the last step, we measured the correlations between those specific items, assessing estimations for other people not from the hospitality industry: (Q3) How do you rate/judge the risk of a fellow countryman in your country of residence being infected with coronavirus? (Q6) How likely is it that a fellow countryman in your country of residence gets vaccinated for coronavirus once a vaccine becomes available on the market? (Q9) What are the chances that a fellow countryman in your country of residence will lose a job because of the coronavirus pandemic? (Q12) What are the chances that the average company in your country of residence will go bankrupt because of the coronavirus pandemic?
Again, we obtained two significant mediocre and positive correlations (0.4 < r < 0.5) between the estimation of the chance that another company in their country will go bankrupt (Q12) and the estimation of the probability that a fellow countryman will lose a job (Q9), as well as between the estimation of a fellow countryman being vaccinated for coronavirus (Q6) and the estimation of the probability that a fellow countryman will lose a job (Q9).
Additionally, we obtained three significant weak and positive correlations (0.1 < r < 0.3). The first was the estimation of the chance of a fellow countryman being vaccinated for coronavirus (Q6) and the estimation of the risk of a fellow countryman being infected (Q3). The second was the estimation of the chance of a fellow countryman being vaccinated for coronavirus (Q6) and the estimation of the chance that another company in their country will go bankrupt (Q12). The third is the estimation of the risk of a fellow countryman being infected (Q3) and the estimation of the probability that a fellow countryman will lose a job (Q9), (see Table 3). With the goal of the article in mind, however, the links between the unrealistic optimism indicators among the various issues investigated in the study appear to be more significant than simple correlations between answers. Thus, the question arises, for example, as to whether the magnitude of unrealistic optimism concerning the possibility of being affected by coronavirus correlates with the magnitude of unrealistic optimism about a person's own workplace, etc.
To verify this, we created a magnitude indicator for unrealistic optimism, which is the difference between the estimation of the likelihood for "fellow countryman/other company in the country" and "myself/my company." Thus, the higher the score (i.e., the difference mentioned above), the higher the unrealistic optimism. Consequently, we calculated four magnitude indicators, such as the risk of being infected, the probability of being vaccinated for coronavirus, the probability of losing a job, and the probability of bankruptcy. We also calculated the correlations between those indicators. As a result, we obtained only one significant, positive, and weak/mediocre correlation (r = 0.33) between the estimation of chances of bankruptcy and the risk of losing a job due to the coronavirus pandemic (see Table 4).

Discussion
Let us start by highlighting an important issue related to the term unrealistic optimism presented in this article. It is clear that workers exhibit unrealistic optimism while assessing their own chances of losing a job and a workplace in comparison to the average HoReCa worker. This biased and illusional thinking supports hypothesis #1. Quite unexpectedly, the pattern of results obtained in our study demonstrated that among our participants, unrealistic optimism was also present when they compared themselves with the general population/average countrymen in terms of the possibility of contracting COVID-19 and job loss as a result of the pandemic. What is even more interesting is that this bias also applies to the entire company that employs the participants, consistently lining up with the same pattern of results. All these results were unexpected and contrary to hypothesis #2, since we had expected that the whole HoReCa industry would feel particularly vulnerable in the time of the COVID-19 pandemic. This pattern of results might be treated not only as a set of positive illusions, but rather as delusion in the light of the objective data. Research on 5819 small businesses (employing 50% of the workforce in the USA) has found that layoffs and closures took place to a large extent [2]. BLS (United States Bureau of Labor Statistics) reported that, as of May 2020, 5.5 million workers from the leisure and hospitality super-sector were suddenly left unemployed as a result of COVID-19.
Taken together, the fact that our participants believe they are less likely to get sick than others, sounds an alarm bell at the very least. Essentially, one should take into account, because of the specificity of their work, that they belong to a high-risk group, due to multiple close interactions with strangers. Despite this fact, however, they consider themselves less exposed to the disease than not only the average person in their business but also than a fellow countryman in their country of residence. Simply put, this bias makes them an even more at risk group since they are not only exposed but also may be less careful with active protection behaviors.
This illusion/bias may influence their salary expectations; they may demand increased spending on their wages [16,37]. As a result, more pressure may be put on business owners. On the one hand, the economy has worsened, and one may expect that customers will not come back in great numbers [2], bills may not be as high as they were, and due to social distancing, fewer guests will be allowed, reducing revenue for the business. A recent study has demonstrated that small companies have saved an average of $10,000, securing operations for only two weeks [2]; restaurants are not any different [39]. On the other hand, workers may demand higher salaries, which, per se, may lead to bankruptcy. The question of how to deal with this clear conflict arises. In all likelihood, the shorter answer is to make workers realistic about their expectations and the risks for the whole HoReCa industry. Otherwise, this specialized hospitality area may be hit hard twice: by COVID-19 and by its repercussions, such as unrealistic optimism.
However, it is also worth considering the other, much more positive edge of the sword's blade. Unrealistic optimism about the possibility of keeping a job and about the future of the company where they work, which was displayed by the majority of participants, seems to be psychologically beneficial, because such a belief helps to reduce stress and anxiety [19,20] and helps the person to retain a sense of personal control [21,40].
Interestingly, there were no differences in judgments about COVID-19 vaccination tendencies. The participants did not declare that they were more likely to be vaccinated in the future than other people. This is especially important from the point of view of hospitality management. Workers in the hospitality industry are not likely to be vaccinated, making eradication of the COVID-19 pandemic impossible. What is more dangerous is that the hospitality industry may be especially vulnerable to future infections, even when a vaccine is available (and in the vast majority of countries, vaccines are not available) especially while facing the delta variant surge. Not only are there a high number of interactions, but also there will be unvaccinated people. Simply put, HoReCa hotspots may become even "hotter". Future interventions should concentrate on reducing this very dangerous threat through global education, to reduce hesitancy toward vaccination.
Regarding the correlation data, they display a weak relationship between certain analyzed variables. A clear relationship is noted only between the answers that are logically related to each other: the professional future of the participant and the future of the company where he or she works. Similarly, it is not surprising that we obtained a significant correlation between the estimated likelihood of getting sick and the declaration of willingness to get vaccinated against COVID-19.
What is especially interesting, however, is the lack of correlation between certain indicators for unrealistic optimism regarding various domains (i.e., between one's own health and own professional future, and between one's own health and the future of the company where the participant works). This means that while unrealistic optimism is common, it is not a personality disposition that activates this bias with the same strength, regardless of the situation or particular life domain. This result is especially hazardous for this sector of business: owners cannot reverse this trend by hiring different workers to reduce this effect. Summing up the results of our research, it can be said that they bring important implications for the entire HoReCa industry, especially for parts of the world where vaccines are not widely available at the time of the delta variant surge.

Recommendations
A recent study on customers' expectations demonstrated that it is expected that the owners provide new technologies for services in the COVID-19 era, to reduce human-tohuman contact [32]. For example, patrons of this business sector expect wireless payments to be employed to reduce physical contact and expand physical distance. In the scope of the aforementioned study, it is clear that new social-psychological-technology should be provided.
From the point of view of the practical implementations of the study, we postulate the need for: (1) wide research on reducing harmful (to the workers) unrealistic optimism bias, and (2) global change in the narrative of COVID-19 management.
First, companies should be equipped with mechanisms leading to the greater safety of their employees. In short, a wide infection rate across the hospitality industry will lead to future closures for many months, and to permanent closures for many businesses. Reversing this trend-by reducing unrealistic optimism-will lead not only to greater safety but also to saving people's jobs, together with the whole industry.
There are some studies that demonstrate how to reach this goal [41][42][43] and even experiments during the COVID-19 pandemic [44], but significantly, they were not conducted in the hospitality industry. Consequently, some aspects may be directly applicable, but more specific studies are needed.
Second, political leaders should change the narrative when speaking of health-related issues of the COVID-19 pandemic, not be too optimistic in terms of the economy of each country, but at least to be more realistic (if not pessimistic). This carefully designed narrative could keep the general population more optimistic in terms of economy (boosting investments, consumption, etc.) while at the same time, make them more actively healthoriented. Refs. [45,46] clearly demonstrate the expectation to change operation practices in hospitality is needed, as well as to change the business orientation to online delivery [47]. On the other hand, future natural disasters, like a pandemic, will not only stay but expand, affecting the hospitality industry to a large extent [48,49]. Our paper delivers potential new solutions to address these needs.
In summary, it is clear that the present study adds new data and recommendations to the body of literature for the hospitality industry to provide new services and protocols [6,46].

Limitations
This preliminary research is not free from limitations. Firstly, this is a preliminary (first) study on this specific group. Not only should it be conducted, for example, in a different country, but also one should keep in mind that, in this research, participants were workers of the restaurants. With the present data in hand, we cannot predict if the same pattern of results would occur among the other branches of the HoReCa industry: catering and hotels. More studies should address this important caveat of the broader picture.
Secondly, our study was run during a period when vaccines were not accessible to the public. This important limitation calls for future research assessing if the availability of vaccines changes the pattern of the results.
Thirdly, since vaccines are available in developed countries and even among these countries the percentage of full immunization varies drastically (in Italy the percentage is very high, in Romania: very low [50]). This paper cannot address this important issue.
Taken altogether, an issue of generalizability is an important limitation of this study.

Institutional Review Board Statement:
This study was reviewed and approved by the ethics committee of the SWPS University of Social Sciences and Humanities (03/P/04/2020).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Conflicts of Interest:
The authors declare no conflict of interest.