The Fear of Covid-19 and Perceived Employability Effect on Emotional Exhaustion and Work Engagement: The Mediating Role of Job Insecurity

Frontline workers are one of the jobs affected due to Covid-19. Covid-19 events can have negative and positive consequences for frontline workers. This study aims to see how fear of covid affects employee emotional exhaustion during a pandemic. Based on the conservation of resources theory, fear of covid-19 can cause frontline workers to experience emotional exhaustion. In addition, during a pandemic, employees can feel high job insecurity, resulting in low work engagement in the company where employees work. The method used in this research is quantitative. Data analysis was performed on 259 company employees who worked as frontline workers using the Structural Equation Modeling technique. The results of the analysis show that fear of covid-19 has a positive effect on job insecurity, and job insecurity has a significant positive effect on emotional exhaustion. In addition, perceived employability has a positive and significant effect on job insecurity and a significant positive effect on work engagement. It is essential for companies to pay attention so that employees do not have high job insecurity and high emotional exhaustion and can create employee work engagement by conducting various kinds of training and health education to increase engagement and reduce employee fear of Covid.

Emotional exhaustion can affect and have a destructive impact on both employees and the organization (Zhou et al., 2020).Suppose emotional exhaustion is a negative consequence for the company's and employees' sustainability.In that case, there are positive consequences for employees, one of which is work engagement, which has a good influence on the welfare of employees as individuals, teams, and work (Bakker, 2017).Work engagement is a fundamental business driver for company success.Employees with work engagement will have vigor, dedication, and absorption as part of their work experience (Lenzi et al., 2021).Frontline workers, who are the face of the company, are directly related to customers, and frontline workers have specific targets that are not only related to economic exchanges but also related to interaction (Hartline et al., 2000;Lages & Piercy, 2012;Guo et al., 2017).Research conducted by Lee et al. (2016) states that work engagement can be affected by job insecurity and negatively affects frontline workers who deal directly with customers.
The current study shows that employees who fear the risks caused by CO-VID-19 will have an impact on job insecurity (Cehn & Eyoun, 2021).Job insecurity asks employees to invest and transfer their resources when individuals feel their resources are being threatened (Greenhalgh & Rosenblatt, 1984).When the exchange of resources is unequal, this will encourage negative consequences such as the inability of employees and finally feeling tired of losing the resources they have.Frontline workers have those targets because the main point of frontline employees is related to customers (Kutaula et al., 2020).Frontline workers struggle to cope with physical and mental demands during COVID-19 (Voorhees et al., 2020) since they directly dealing with customers, especially as the bridge between companies/ organizations and potential customers (Chatterjee et al., 2022).In the context of the pandemic caused by COVID-19, employees may receive negative consequences from various dimensions, both in life and at work.
Two years after the post-COVID-19 pandemic, the employment perspective of young people remains uncertain and has increased in terms of their vulnerability.Self-perceived employability, which refers to an individual's self-perception of the possibility of getting a job in the labor market, reduces an individual's fear of being unemployed while increasing feelings of being in control of one's work life (Lechien et al., 2022).Demonstrating the effects of self-perceived perceived employability in the context of a pandemic will provide employers and employees with insight into how to survive the negative impacts of pandemics and similar situations in the future.This research has exciting novelties to study, which are using respondents as frontline employees, especially the majo-rity in the service industry in the banking and hotel sectors, working onsite in Indonesia.in the context of the pandemic and also to fill the gap between how respondents react to job insecurity with positive and negative behavior.

Hypothesis Development
Fear of COVID-19 refers to unpleasant emotions caused by perceiving a perceived threat (Pakpour & Griffiths, 2020).The Conservation of Resources (COR) theory, owned by (Hobfoll, 1989), is one of the theories that underlie this hypothesis.Hobfoll (1998), with the Hobfoll COR theory, can explain the understanding of the fear of covid-19 and emotional exhaustion (Hobfoll, 1989).Previous studies also stated that losing their resources would cause emotional exhaustion because individuals may perceive themselves as no longer having sufficient resources to deal with their stress (Hobfoll, 1989).So, based on the description above, the researcher hypothesizes that: H1: Fear of COVID-19 has a positive effect on emotional exhaustion.
The fear of COVID-19 and the several lockdowns have created particular fears, especially about job insecurity among employees across various industries (Abbas et al., 2021).The psychological contract theory is used to understand the causes of job insecurity owned by employees (Keim et al., 2014).Job insecurity can be a significant issue during a pandemic.Study in US by Ganson et al. (2021) show that job loss will affect employees' mental health.Previous studies have also shown that uncertain environmental factors such as health crises, economic downturns, technological changes, and political uncertainties will cause employees to feel insecure about their jobs (Lee & Jeong, 2017).Environmental factors such as high unemployment rates are also associated with job insecurity due to a lack of control during uncertain times (Debus et al., 2012).H2: Fear of Covid-19 has a positive effect on job insecurity.
Conservation of resources theory and job demand resources are the theoretical basis.During the COVID-19 period, perceived employability possessed by individuals will become a source of potential, resulting in individuals losing their resources and emotional exhaustion.
Previous studies also stated that job insecurity could make individuals lose their resources, including emotional exhaustion, depression, and anxiety (Piccoli & De Witte, 2015).Based on the explanation above, the hypothesis proposed is: H3: Job insecurity has a significant positive effect on emotional exhaustion.
Work engagement is defined as a positive, satisfying, work-related state of mind that is characterized by enthusiasm, dedication, and absorption in work (Karatepe, 2013).In perceived employability, there is a side of self-reliance (self-confidence) that is used by an individual to survive and want to be successful in the future so that it will impact the work environment (Demerouti & Bakker, 2011).As part of personal resources, perceived employability positively affects work engagement.Empirically, research conducted by Cuyper et al. (2008) said that there is a positive influence from perceived employability and work engagement H4: Perceived employability has a positive effect on work engagement.Dekker & Schaufeli (1995) states that employees with a lower academic level are more prone to job insecurity than highly qualified workers.That statement is because someone with higher education has high perceived employability.In this case, employment security is conditioned by the ability of employees to work (Folkman et al., 1984).They concluded that employable employees tend to face less job insecurity.H5: Perceived employability has a negative effect on job insecurity.
Some previous literature states that job insecurity will reduce employee work engagement.This phenomenon is carried out and researched in a variety of different contexts: Italian workers (Guarnaccia et al., 2018); expatriate employees from various companies in Saudi Arabia (Ali et al., 2017) and international hotel employees in China (Karatepe et al., 2020).Stander & Rothman (2010) also stated that employees who feel their jobs have a high level of insecurity will have low engagement.Thus, the following hypothesis, therefore: H6 : Job insecurity has a negative effect on work engagement.
Based on the COR theory, individuals will experience emotional exhaustion when they lose resources and have no more resources (Hobfoll, 1989).Job insecurity asks employees to invest and transfer their resources when individuals feel their resources are being threatened (Greenhalgh & Rosenblatt, 1984).When the exchange of resources is unequal, this will encourage negative consequences such as the inability of employees and finally feeling tired of losing the resources they have.Those feelings can encourage employees to invest extra effort at work to maintain their existing positions.So, the hypothesis that can be proposed is: H7: Job insecurity mediates the effect of fear of COVID-19 on emotional exhaustion.
Van Vuuen et al. (1991) and De Witte (1999) state that job insecurity acts as a stressor that will later be able to have a negative effect on employee work engagement.When employees feel that they are insecure at their jobs the employee will feel they have no more power and will feel they have less control (Bosman, 2005;Vander Elst et al., 2013).H8: Job insecurity mediates the effect of perceived employability on work engagement Based on the explanation of the hypothesis development above, it can be seen that in Figure 1 about the hypothesis framework.

MeTHOD
The method used in this study is the quantitative method.This research uses quantitative methods.The study in this research used a cross-sectional design.According to Sekaran (2013), a cross-sectional design is a research design with a data collection process that is only carried out once, periodically, for days, weeks, or months to answer research questions.This research was conducted using the structural equation modeling (SEM) method.This study used primary data in the form of a self-administered questionnaire.This study gathers data from 259 employees.The frontline workers who are the research sample are frontline workers who work to provide services, especially in the service industry, such as hotels and banking.From the research results, the age range is 25-55 years.Frontline workers also have the skills to build good customer relationships and provide services to them.The SEM to be used is LISREL version 8.8.Data processing in this study was carried out using Lisrel because it can simultaneously measure latent variables and indicators (first order).Also, Lisrel can test latent variables that contain sub-dimensions of other latent variables (second order).
A self-administered questionnaire is one form of data collection in the form of questionnaires that the respondents themselves answer.Some sources of secondary data are statistical bulletins, government publications, published and unpublished information available inside or outside organizations, company websites, and the internet.The research sample used is frontline employees.The unit of analysis refers to the level of data aggregation collected during the data analysis stage (Sekaran, 2013).The mediation test is a test to influence the indirect effect of a variable.The mediation test used in this study regarding the analysis of mediation tests with a mediation decision-making chart.Zhao et al. (2010) explained that decisionmaking was taken from Baron & Kenny (1986) who divided the types of mediation into complementary, competitive, full, and partial mediation.
This study uses five variables, and each variable has several research indicators.The fear of COVID-19 variable was analyzed using a 7-item scale adopted by Ahorsu et

Result
Respondents are frontline employees given pre-screening questions to ensure the criteria are met.From the analysis of the respondent's profiles, 127 respondents were male, and the remaining 132 respondents were female respondents.Age less than 25 years dominated the respondents, with a total of almost more than half of the respondents, namely around 131 respondents that age.Undergraduate education also dominated with more than half of the respondents, followed by high school education level.There were 120 permanent and 139 contract employees who participated in this study, and various types of jobs ranging from sales staff to restaurant servers were included in the survey.
Table 1 is helpful in explaining the validity of all variables.Validity is used to see whether all indicators are valid for use in research.
In this study, all the items in the questionnaire are valid.In Table 1, the validity test of the fear of COVID-19 explained that the loading factor value for the fear of COVID variable is 0.60-0.89,and all items are valid.Meanwhile, all measurement indicators for job insecurity variables have a t-value ≥ 1.96 and have a standardized loading factor or standardized validity value of SLF ≥ 0.5.The emotional exhaustion variable has a t-value ≥ 1.96 and a loading factor value ≥ 0.5, ranging from 0.68-0.90.The perceived employability variable has an SLF value of 0.82-0.88.All dimensions of the valid work engagement variable are listed in Table 2. Vigor, dedication, and absorption are valid measurements.All dimensions follow the reference indicators that have SLF values ≥0.5.Based on Table 1 below, it can be seen that the t-value of all variable indicators has a value of > 1.96.Meanwhile, the standard loading factor value results have a value of > 0.5. 1 is that all indicators are valid and reliable.For fear of COVID-19, the most significant loading factor that is owned is in points 5 and 6, which have the same value of 0.89.The most significant loading factor on the emotional exhaustion indicator is point 4, which states that working with other people is tiring.On the job insecurity indicator, the most significant loading factor is in point 3, which explains that frontline employees have concerns about their jobs.

So, the conclusion from Table
Indicators Composite Reliability and Average Variance Extracted (AVE) to test the reliability.The standard value of CR is ≥ 0.7, and the average variance extracted (AVE) value ≥ 0.5 is used in this study.The overall results show that each variable's CR and AVE values have a CR value of ≥ 0.7 and an average variance extracted (AVE) value of ≥ 0.5.Thus, all variables can be said to be reliable.Based on the results of the structural model test, it can be seen that for the absolute fit model test, there are three indicators with measurement results indicating poor fit, including the p-value, SRMS, and GFI.Meanwhile, RMSEA has a marginal fit.A good calculated value, with the conclusion that good fit is indicated by NFI, NNFI, CFI, IFI, and RFI, which have a calculated value of ≥ 0.09.
The criteria for how much significance is indicated by the t value on the variable in the amount of ±1.645 with the acceptance rate of the hypothesis as much From the results of calculating the research hypothesis in Table 3, it can be seen that the t-value is above 1,645 (t-value ≥1,645) for hypothesis 1 with a t-value of 1.69 and a coefficient of 0.10, the effect of fear of covid-19 on emotional exhaustion has a positive and significant effect.Furthermore, hypothesis 2, the effect of fear of COVID-19 on job insecurity, has a significant positive effect with a t-value of 8.72 and a coefficient of 0.55, so hypothesis 2 can be accepted.Hypothesis 3 is also accepted with a t-value of 9.82 and a coefficient of 0.66, which significantly positively affects job insecurity and emotional exhaustion.
In hypothesis 4, it can be seen that the t-value is more than 1.645, so the hypothesis is accepted and is positively significant.Hypothesis 5 indicated that the effect of perceived employability toward job insecurity has a significant negative effect.Hypothesis 6 states that the effect of job insecurity on work engagement has a significant negative effect.In hypothesis mediation, in hypothesis 7, from Table 4, it can be seen that the indirect effect (a x b) has a value of 0.363, which means that it has a significant positive effect.Hypothesis 8 represents that the indirect effect (a x b) has a significant negative effect while the direct effect (path c) has a significant positive effect.Figures 2 and 3 illustrate the path analysis of the hypotheses mentioned.In Figure 2 shows a path analysis of 8 hypotheses, which is reflected in Tables 2 and 3. .Figure 3 shows the results of the Lisrel output used as a research analysis tool.

Discussion
Fear of COVID-19 is an unpleasant emotion experienced by some employees and can damage/obstruct their ability to think (Pakpour & Griffiths, 2020) .Greenberg et al. (1986) explained that the conservation of resources theory from Hobfoll (1989) states that when fear occurs due to an infectious disease, for example, the fear of being infected with the coronavirus during a pandemic, can reduce individuals' performance in working, and increase their emotional exhaustion.
This statement is also in line with research conducted by López et al. (2020) which explains that the COVID-19 pandemic can cause burnout syndrome, as well known that emotional exhaustion is one of the signs of burnout.Conceptually, fear is an antecedent of non-adaptive cognitive change mechanisms that will cause disintegration in individuals with detrimental effects (Khudaykulova et al., 2022).Individuals will experience greater job insecurity because of fear of losing (Moore et al., 2004).Research conducted by Toros et al., (2022) also states that fear of covid-19 positively affects employee job insecurity.Job insecurity will reflect an individual's fear of their future jobs (losing their current job) (Sverke et al., 2002).Several previous studies support that job insecurity will cause a person to feel burnout and reduce employee well-being (Dekker & Schaufeli, 1995).
Conservation of resources theory explains that employees have multiple resources, such as time and energy, and they attempt to bring them back when it is lost (Hobfoll, 1989).Empirically, research conducted by De Cuyper et al. (2008) said that there was a positive effect from perceived employability and work engagement.Employability has been considered to give individuals a sense of autonomy and make them more likely to achieve workplace participation.So, employees with high employability will also have high work engagement.Perceived employability and job insecurity often deliver nega-tive effects.The reality of both problems commonly refers to individuals'subjective experiences (Kinnunen et al., 2014).Even though the level of perceived employability of individuals has increased in the context of a pandemic, job insecurity has also increased .Job insecurity during a pandemic will result in an individual's inability to cope with job security in the future.Ornell et al. (2020) also argue that the COVID-19 pandemic can cause insecurity at work.In Hypothesis 5, it is explained that the perceived employability of frontline employees has a significant negative effect.Based on Alpler & Arasli (2020) state that one of the antecedents that have the potential to cause job insecurity is perceived employability.Other research shows different results.It was found that perceived employability has a positive significant effect on job insecurity.This result indicates that perceived employability during a pandemic still tends to be high.
Contextual factors, such as the availability of jobs related to supply and demand, impact on employee perceptions (Forrier et al., 2015).The conservation of resources theory underlies some explanation about the mediation between fear of COVID-19 and emotional exhaustion on job insecurity.This theory explains that individuals' resource problems are caused by a threat to how to preserve their resources they have (De Cuyper, 2014).As in the hypothesis explaining perceived employability on work engagement, it is known that individuals with a high level of perceived employability tend to have diffe-

CONClUsION AND ReCOMMeNDATION
Fear of COVID-19 has a significant positive effect, and job insecurity is the mediating variable of this study.Job insecurity is a mediating variable that explains that when an employee experiences fear of COVID-19, the employee must first feel job insecurity so that it has an impact on the employee's emotional exhaustion.Perceivedemployability in this study has a significant positive effect on work engagement and a significant negative effect on job insecurity.The effect indicates that when perceived employability is high, the job insecurity felt by employees is also high.Perceived employability has a significant positive effect on work engagement, which explains that when employees feel high perceived employability, employee work engagement is also high.Research that will be carried out further can be carried out with a more extensive and comprehensive sample to represent the research results obtained.In addition, from an industrial standpoint, the research subjects can also be differentiated so that the results of further research can explain how the comparisons are made in each industry.Differences in research subjects other than frontline employees can also add to the diversity of research results.
Research using the longitudinal or daily diary method so that the re-sults obtained are also more accurate because there is a variable of emotional exhaustion,which in this study is recommended to be carried out with a longitudinal study.Longitudinal methods and qualitative research can be used for further research so that results can increase validity and draw better conclusions.Additions or differences in the variables studied are also suggested in this study, one of which is the existence of a moderating variable.In the future, further research can add several moderating variables that will create more comprehensive results.

Table 1 .
Validity Test First Order

Table 2 .
Validity and Reliability Test Second Order

Table 3 .
Research Hypothesis Test

Table 4 .
Mediation Hypothesis Test The Results of Full Model Analysis Lisrel rent qualifications, so they are easy to get hired.Individuals have several obstacles to achieving work engagement (Suleman et al., 2021), so they need to readjust to the context and conditions of employee engagement.Table 2 helps explain explaining the validity of all indicators.Validity is used to see whether all indicators are valid for use in research.From the table below, vigor and dedication have the most significant SLF value.