Staying at Work? The Impact of Social Support on the Perception of the COVID-19 Epidemic and the Mediated Moderating Effect of Career Resilience in Tourism

Abstract

The COVID-19 epidemic has caused dramatic impacts and changes in the tourism industry, and job insecurity and emotional exhaustion have created psychological stress and negative emotions. Social support for Taiwan tourism workers (travel agency, transportation industry, lodging industry, tourism and leisure industry, etc.) plays an important role in their career resilience. However, not all of the potential social support moderators have a critical impact. This study used PLS-SEM analysis to survey 373 respondents by using an online questionnaire to investigate the critical influence of social support on the spread of COVID-19 using career motivation theory. In addition to the direct relationship between the individual’s psychological resilience and social support, the strategy of social support (family and friends, national relief policies and workplace support) is also pointed out. The results of the study illustrate the effectiveness of workplace support in combating the epidemic. This study provides information on effective resistance to the epidemic, how to prolong career resilience during unexpected shocks and stresses, and how to understand the mechanisms of adaptation or resilience in adversity and complements the study of factors and literature base in resilience research. It is also used as a study of the impact factors and industry strategy planning in future research.

1. Introduction

The COVID-19 outbreak and its continued spread, according to the United Nations World Tourism Organization [1], showed an 85% decline in the international tourist population from January to May 2021 compared to the same period in 2019, and a 65% decline compared to 2020. International tourist arrivals are down by approximately 147 million compared to 2020 and 460 million compared to 2019, causing a decline in the global tourist population and devastating the tourism industry and global economy. As of 24 January 2022 [2], the outbreak has spread and expanded, infecting 349,641,119 people worldwide and causing 5,592,266 deaths. On 24 January 2022, the World Health Organization classified COVID-19 as a global pandemic.

Since the outbreak of the global epidemic, the goals and directions of the tourism industry have been revised and adjusted downward several times. The tourism industry has suffered the most serious crisis due to the prolongation of the epidemic. It is estimated that it will take two and a half to four years to return to 2019 levels [1]. The tourism industry is a labor-intensive industry that requires time to develop human resources. Therefore, it is more important for business managers and employees to be able to respond to the spread of the epidemic and to recover the economy. After this outbreak, long-term resistance to the epidemic and organizational flexibility is key to the industry’s survival. The global pandemic of COVID-19, especially in Asia, has been spreading particularly early and the industry has been seriously affected. Up to 49 million jobs will be at risk across the region, equivalent to a loss of nearly $800 billion in tourism and tourism domestic production [3]. Since the outbreak of the epidemic, Taiwan’s tourism industry has been facing a severe impact on tourism. Initially, direct flights were boarded and quarantined, then entry was controlled, and finally, foreigners were subjected to a comprehensive entry control policy. This resulted in setback, depressed environment, and stagnation in the domestic and international tourism, lodging, and air/cruise transportation industries [4,5,6,7,8]. Afterwards, organizations faced the same harsh facts of layoffs and bankruptcy as the rest of the world [9,10].

Countries around the world are aiming to promote sustainable tourism and social well-being to revitalize global tourism [7,8,11]. The Taiwan government has proposed to introduce financing loans and interest subsidies, fuel or license tax subsidies for land, sea, and air transportation in the transportation industry, respectively. For the tourism industry, the government has implemented relief programs such as subsidies for stopping inbound and outbound tours, relief for inbound travel agencies, subsidies for human resources training, and interest subsidies for financing revolving loans [12]. Most of the countries have proposed solutions in terms of “financing”, “employment”, and “taxation”. For example, the French government provided 18 billion euros to support the recovery of the tourism industry and EUR 6.2 billion in loans to 50,000 companies. Denmark established a EUR 1.5 billion guarantee fund to compensate travel agencies for refunds related to canceled orders. United States invested USD 2 trillion to revitalize the tourism industry [13]. Many critical issues should be of concern, such as preparation of labor services and market policies during other outbreaks [14], the role of government in revitalizing tourism [15], how to make use of the industry toughness of the tourism industry [16,17], and how practitioners adapt and adjust to the difficult situation and face the impact of the long-term fight against the epidemic.

Moreover, concerning psychological stress and negative emotions caused by job insecurity and emotional exhaustion in the face of a raging epidemic [18,19,20,21], exploring the relationship model of career motivation theory [22], understanding the key influences of social support in the spread of COVID-19, in addition to the direct relationship between individuals’ psychological resilience and social support [23,24], all are important to further explore the ability of social support to positively enhance individuals’ career resilience [25,26,27,28].

In addition, this study incorporates government support, such as loans, subsidies, or tax breaks, combined with the psychological idea of positive thinking [29,30,31] to further explore whether state resources, support from family and friends or organizations can be adapted to strengthen the resilience of individuals in their careers as an adaptation mechanism under the epidemic.

2. Literature Review and Hypothesis Development

2.1. COVID-19 Impact Relationship Study

The epidemic (COVID-19) is raging around the world and is having a significant impact on people’s physical and psychological well-being [32]. Psychological concerns, including anxiety, depression, fear, and perceived risk, have a significant impact on life satisfaction [33,34,35,36,37,38]. The rapid spread and expansion of COVID-19 have created a global medical crisis, increasing fear and anxiety. The quality of sleep of health care workers at the beginning of the outbreak was indeed affected by anxiety, stress, and self-efficacy [39,40,41]. In addition, pregnant women feel vulnerable and susceptible to the spread of the epidemic and are affected by stress and fearful perceptions [42,43].

In addition to the psychological impact, the epidemic included changes in people’s perceptions and behaviors, such as the perceived severity of vulnerability during the epidemic, which had an impact on the intentions followed. Exploring the application of the Theory of Planned Behavior to protect behaviors and subjective norms for preventive behaviors when people are exposed to epidemic threats [44], or changes in consumer behavior and intentions after the epidemic [45]. Country-specific guidelines for epidemic preparedness, appropriate risk perception and knowledge of epidemics [46], health belief models for perceived risk, and subsequent health promotion behaviors during epidemic preparedness (i.e., ‘inactive tourism’) are important [47]). As the global epidemic spreads, the population moves rapidly and the travel space becomes more open, healthy and safe travel behavior becomes more important. Tourism, leisure and recreation, air transportation, and lodging industries are all facing uncharted challenges in responding to and changing visitor behavior and industry operations.

The indefinite spread of COVID-19 quickly shifted the global tourism issue from a focus on how to manage marketing to air border control [48,49,50], cruise shipping control [10,51,52], hotel closing [53,54], strategies for disruptions and closures in response to the suspension of the opening of sightseeing spots [55], and proposals for the economic recovery of the tourism industry after the epidemic [5,8,15,56,57,58,59]. The epidemic has created a strong and persistent dilemma for the global tourism industry. Most of the other epidemic-related studies focus on global economic recovery, industrial development trends, and corporate transformation strategies. From the industrial problems brought about by the epidemic to the recovery and transformation after the epidemic, this study focuses on individual career resilience to explore the career pressure, crisis, and adjustment problems brought about by the epidemic.

Hypothesis 1 (H1).

The perception of COVID-19 will have a significant positive impact on career resilience.

2.2. Social Support

With the spread of the epidemic, social support is important for the industry during recession, constraints, and changes. Social support for stress originates from the field of clinical psychology, which explores the interaction between human behavior and the environment. It provides interpersonal networks of psychological and material resources that enable individuals to cope with stress [60]. Social support is also a cognitive protective factor for individuals to recover from despair in the face of stress, frustration, grief, or hardship [61,62]. People who perceive higher levels of social support actively seek out the resources needed to solve their problems and avoid negative attributions and thoughts [63,64].

The effect of perceived social support on mental health and anxiety has been demonstrated in different types of crises [65,66]. Social support is important for the social development and mental health of individuals and can be divided into two parts: objective and subjective [62,67]. The objective component refers to the actual social support (government grants and subsidies) received during the outbreak. The subjective component refers to perceived social support (expectations and perceptions of social support), such as timely parental support and assistance to help adolescents cope with the fears associated with COVID-19 [68,69]. Then, social bonding for seniors can alleviate negative psychological thoughts [70]. Social, family, and friend support during the epidemic can reduce depression and improve the mental health of pregnant women [71]. On the other hand, objective support refers to support for organizing unions or government subsidy measures during an epidemic [72,73,74]. The subsidy measures for Taiwan’s tourism industry (tourism, transportation, private tourism and entertainment, public tourism and leisure, etc.) are shown in Figure 1 below. With the uncertainty and instability of the global tourism environment, family, peers, friends, and government support play an important role and social support also has an important influence. Most of the previous studies focused on the relationship between employees’ life satisfaction and loyalty. However, this study will examine the relationship between the mediating role of social support and career resilience of tourism workers.

With the uncertainty and instability of the global tourism industry environment, family, peers, friends, and government support play an important role. Most previous studies have focused on the relationship between employee life satisfaction and loyalty, organizational commitment, and organizational justice, but have not yet analyzed the relationship between social support influences in times of crisis. Therefore, this study will examine the relationship between the role of social support mediators and career resilience of tourism workers (travel industry, transportation industry, private tourism and recreation industry, public tourism and recreation industry, etc.) in the context of unexpected stressful events (COVID-19).

During the COVID-19 period, social support formed an important stress buffer [75], and those working in the tourism industry were provided with social support to express their emotions and resources to mitigate the stress on their physical and mental health [76,77,78]. Organizational social support in a difficult and unstable environment during COVID-19 can reduce job insecurity among hotel employees [19,20]. It would positively adjust mental health, self-esteem, economy, self-efficacy, etc. [79,80,81]. By providing a good adjustment mechanism through social networks, the higher the employee’s occupational adaptability, the higher the perceived support from supervisors or the organization, the higher the positive impact on work attendance and performance [82,83], the higher the organizational loyalty [84,85], and the lower the willingness to leave [18,20,28], lower turnover intentions [26,62], or recovery from uncertainty and despair. Thus, social support provided a positive moderating effect on the stress of business closures, employee furloughs and layoffs, and changes in business models and operations during COVID-19.

Hypothesis 2 (H2).

The perception of COVID-19 will have a significant positive impact on social support.

2.3. Resilience and Adaptability

Resilience helps individuals to resist adversity and to adapt and develop well in difficult situations [86]. In particular, it plays an important role in times of loss, trauma, and adversity [87], and resilience is a dynamic learning process that interacts over time with the ability to “bounce back” from negative emotional experiences of adversity, uncertainty, and threat [88,89]. When applied to the workplace, resilience allows individuals to face workplace adversity without being crushed by setbacks, stresses, and threats, and to actively restore positive emotions or make positive career adjustments [90,91]. Resilience protects people from the negative and stressful events of daily life and increases an individual’s ability to cope with potential threats, and resilience research has examined the relationship between life satisfaction [86,92]. Happiness [93], optimism in psychological capital, and self-efficacy [94] can reduce the negative effects of career traumatic events such as depression, anxiety, or job insecurity in terms of career resilience and adaptation [41,95]. The psych emotional aspects are based on occupational adaptability [96], turnover intention [28], social support [27], and mindfulness [97] were explored to strengthen career resilience. Resilience in the career can be considered as a defense mechanism that allows people to thrive in the face of adversity while increasing resilience strengthens career flexibility.

Career resilience comes from London’s career motivation theory, which suggests that career resilience is the key to career motivation and is the ability of an individual to cope with and adapt to the work environment. The motivation to be resilient comes from the concept of motivation including self-efficacy, achievement, and patience. Employees with career resilience tend to engage in more effective career management behaviors [22,98]. When developing theories of occupational adaptation and developmental relationships for empirical assessment, Savickas’ [99] occupational construct theory model was mostly applied. There are four conceptual indicators, combined with attitudinal or behavioral scale assessments as a theoretical basis and conceptual extension, combining attitude or behavior scale assessment [96,100,101,102,103,104]. Extending self-efficacy and career beliefs of positive emotions with the concept of four indicators, self-efficacy demonstrates resilience in the face of adversity by activating emotional, motivational, and behavioral mechanisms to promote resilience in stressful situations [105,106,107,108]. Positive emotions help highly resilient people to recover from daily stress [109]. Therefore, individuals with greater psychological resilience in the face of long-term adversity against the epidemic can recover and recreate, and adapt themselves to break through the crisis.

Long-term performance is based on positive resilience, and many studies show that social support has a positive effect on career resilience [23,24] and has a direct and buffering effect [25,27,28]. Social support is an effective mechanism for reducing the effects of depression and strengthening psychological resilience [26]. In summary, there are considerable benefits to studying the use of resilience in a career. During the COVID-19 period, if employees in the tourism industry have the support and input of positive family members, organizations, or related resources in times of stress or adversity, they can strengthen their career resilience and performance, which will help their career endurance and resilience in the industry.

Hypothesis 3 (H3).

Social support will have a significant positive impact on career resilience.

Based on career motivation theory, we hypothesized that perceived COVID-19 effects would have a positive effect on career resilience and that social support would moderate the relationship between perceived career resilience through a relationship model and social support mediators. Next, after understanding its mediating role, the key influences of social support will be further explored. In the context of COVID-19, this study conducted a questionnaire survey of practitioners in the tourism and leisure industry (travel industry, tourism transportation industry, and lodging industry, etc.) to investigate the influence mechanisms in the career resilience model.

3. Methodology

3.1. Sampling and Population

During the epidemic period, in order to avoid contact infection, the study conducted an online systematic questionnaire survey from March 2021 to April 2021. The samples were employees of Taiwan’s tourism industry (such as travel agency, tourism transportation, accommodation, exhibition industry, public/private tourism, and leisure industries). In order to ensure the effectiveness of the survey, the questionnaire is divided into three parts. The first part first confirms the occupational categories of the sample, the second part is the work experience survey, and then starts the questionnaire. Sampling was collected through e-mail, open community (Facebook), and social media software (Line, What App). A total of 373 valid questionnaires were collected. The questionnaire was measured on a five-point Likert scale, and scores were given from 1 to 5 (1 = strongly disagree, 5 = strongly agree) in order of agreement.

The scale was designed concerning Fourie and VanVuuren’s [110] Occupational Resilience Scale (CRQ for short). The scale was developed based on career motivation theory and combined with the Multidimensional Scale of Perceived Social Support (MSPSS). It was used to understand respondents’ perceptions of external support [111]. In this study, the reliability and validity of the PLS-SEM model were analyzed and explained to verify the explanatory and predictive power of the model’s path coefficients, and the most approximate method of model application was measured to verify the convergent and discriminant validity of the model. PLS-SEM has advantages in terms of estimation robustness and statistical power for small samples [112]. Compared to other methods, PLS can be applied to complex structural equation models with a recommended sample size range from 30 to 100 cases [113]. If the sample size is 200 or more, the investigator can reasonably study the characteristics of the potential variable model [114].

The research design is divided into three parts. Demographic variables (i.e., gender, education, occupation in the tourism industry, age, job nature, years of experience, and place of residence) are included in 7 questions. The first part, “perceived impact of the epidemic”, was divided into three dimensions (negative emotion, work stress, and positive), and a total of 18 questions were asked to explore the interaction between negative emotion or income reduction and career resilience under the impact of the epidemic; the Section 2.2, “social support” was divided into three components (national support, family and friends support, and career support) and consisted of 12 questions. Finally, the “career resilience” scale is divided into three dimensions (self-efficacy, personal beliefs, and adapting to changes in the career) with a total of 15 questions. This study will investigate the relationship between the components of “epidemic perception”, “social support”, and “career resilience” to understand the changes and adjustments in government policies, career changes, job beliefs, and adaptability during the epidemic.

3.2. Sampling and Population

As shown in

Table 1. Data list of the APS database.

Variable	Number	%
Gender
Male	183	49.1
Female	190	50.9
Age
18~25 years old	6	1.6
26~35 years old	39	10.5
36~45 years old	82	22.0
46~55 years old	151	40.5
56~65 years old	77	20.6
Over 66 years old	18	4.8
Education
Master or above	163	43.7
Colleges and Universities	141	37.8
High School	64	17.2
Junior High or below	5	1.3
Occupation
Travel agency	105	28.2
Tourism transportation	27	7.2
Accommodation	62	16.6
Private Tourism Recreation Industry	17	4.6
Public Tourism and Leisure	31	8.3
Exhibition Industry	6	1.6
Sightseeing leisure recreation educator	83	22.3
Restaurant	31	8.3
Night Market	11	2.9
Job Nature
Entrepreneurs, bosses, heads of agencies	86	23.0
Supervisory Level	103	27.6
Full-time employees	129	34.6
Part-Time Staff	32	8.6
Fleet Staff	23	6.2
Years of experience
Less than 1 year	13	3.5
1–5 years	65	17.4
6–10 years	62	16.6
More than 11 years	233	62.5

Note: Data source from this study.

, 50.9% of the respondents were female, while 49.1% were male; they were aged between 46 and 55 (40.5%); The majority of respondents’ education level was tertiary or above (81.5%); tourism accounted for 28.2% of the study sample; the majority of the sample were full-time employees (34.6%), and 62.5% had more than 11 years of working experience.

3.3. Measurement Model

This study was conducted to determine the resilience, response, and adjustment of tourism and leisure travelers under the spread of the COVID-19 epidemic. Based on the Structural Equation Model (SEM) analysis by Anderson and Gerbing [115], the first stage of the Measurement Model and the second stage of the Structural model were evaluated. In this study, after understanding the reliability of the questions through the validated factor analysis, a hypothesis testing path analysis of PLS-SEM was conducted. The main benefit of using PLS-SEM is that the results are not affected by the small sample size [116]. In this study, Smart PLS 3.2.9 was used to analyze the data, and the results of the Bollen-Stine Bootstrap modified model fit are as follows.

3.3.1. Convergent Validity

The CFA (Confirmatory Factor Analysis) was used to evaluate the measurement model variables, and the reduction in the study questions was corrected according to Kline [117] until the measurement model fit was acceptable, and then the SEM model of the system was constructed. In this study, we applied PLS-SEM to analyze and explain the reliability and validity of the model, verified the explanatory and predictive power of the model’s path coefficients, and measured the model’s convergent and discriminant validity by applying the most approximate method. According to Hair, Anderson, Tatham, and Black [118]; Nunnally and Bernstein [119]; and Fornell and Larcker [120], the convergent validity should be consistent with a standardized factor loading of greater than 0.5. Results of analyses with Composite Reliability (CR) greater than 0.6 and Average Variance Extracted (AVE) greater than 0.5. The results of this study are shown in

Table 2. Reliability and validity analysis of the first-order model.

Construct	Item	Significance of Estimated Parameters		Item Reliability		Construct Reliability	Convergence Validity
		Unstd. Factor Loadings	p-Value	Std. Factor Loadings	SMC 1	CR 2	AVE 3
NE (Negative emotions)	NE1	1.000		0.790	0.624	0.922	0.629
	NE2	1.092	0.000	0.854	0.729
	NE3	1.128	0.000	0.864	0.746
	NE4	1.062	0.000	0.800	0.640
	NE5	0.926	0.000	0.704	0.496
	NE6	1.025	0.000	0.776	0.602
	NE7	1.019	0.000	0.750	0.562
WP (Working pressure)	WP1	1.000		0.658	0.433	0.895	0.516
	WP2	1.298	0.000	0.749	0.561
	WP3	1.254	0.000	0.759	0.576
	WP4	1.327	0.000	0.759	0.576
	WP5	1.176	0.000	0.722	0.521
	WP6	1.294	0.000	0.763	0.582
	WP7	1.049	0.000	0.697	0.486
	WP8	0.964	0.000	0.625	0.391
SG (Positive Self growth)	SG1	1.000		0.778	0.605	0.858	0.664
	SG2	0.970	0.000	0.772	0.596
	SG3	1.129	0.000	0.890	0.792
FAF (Help from family and friends)	FAF1	1.000		0.924	0.854	0.921	0.748
	FAF2	0.906	0.000	0.706	0.498
	FAF3	0.981	0.000	0.875	0.766
	FAF4	1.048	0.000	0.934	0.872
RES (National rescue policy)	RES1	1.000		0.892	0.796	0.929	0.766
	RES2	1.020	0.000	0.906	0.821
	RES3	1.021	0.000	0.894	0.799
	RES4	0.930	0.000	0.806	0.650
WS (Career support)	WS1	1.000		0.630	0.397	0.849	0.590
	WS2	1.018	0.000	0.681	0.464
	WS3	1.184	0.000	0.876	0.767
	WS4	1.135	0.000	0.855	0.731
SE (Self-efficacy)	SE1	1.000		0.836	0.699	0.894	0.584
	SE 2	0.836	0.000	0.700	0.490
	SE 3	0.830	0.000	0.800	0.640
	SE 4	0.813	0.000	0.769	0.591
	SE 5	0.870	0.000	0.771	0.594
	SE 6	0.954	0.000	0.699	0.489
PB (Professional belief)	PB1	1.000		0.660	0.436	0.773	0.533
	PB2	1.264	0.000	0.804	0.646
	PB3	0.980	0.000	0.719	0.517
AC (Accepting change)	AC1	1.000		0.665	0.442	0.913	0.641
	AC2	1.118	0.000	0.779	0.607
	AC3	1.254	0.000	0.869	0.755
	AC4	1.215	0.000	0.888	0.789
	AC5	1.166	0.000	0.897	0.805
	AC6	0.946	0.000	0.669	0.448

Note: ¹ SMC = Squared Multiple Correlations, ² CR = Composite reliability, ³ AVE = Average Variance Extracted.

. The standardized factor loadings were all greater than 0.625; the item composition reliability was above 0.773; the average variance extraction amounted to above 0.516, where items below the threshold limit were removed for the next analysis.

When multiple indicators measure a given construct and thus form a structural pattern, the convergence validity of the construct is more important. It can be extracted by the average variance.

Table 3. Table of reliability and validity analysis of the two-order model.

Construct	Item	Significance of Estimated Parameters				Item Reliability		Construct Reliability	Convergence Validity
		Unstd. Factor Loadings	S.E.	Unstd./S.E.	p-Value	Std. Factor Loadings	SMC 1	CR 2	AVE 3
COVID-19 PE (Perception of COVID-19)	Negative emotions	1.000				0.807	0.651	0.829	0.624
	Working pressure	0.958	0.097	9.860	0.000	0.848	0.719
	Positive self-growth	0.803	0.095	8.472	0.000	0.708	0.501
SS (Social support)	Help from family and friends	1.000				0.653	0.426	0.754	0.507
	National rescue policy	1.094	0.123	8.873	0.000	0.609	0.371
	Career support	1.036	0.123	8.404	0.000	0.856	0.733
CR (Career resilient)	Self-efficacy	1.000				0.815	0.664	0.899	0.749
	Professional belief	0.719	0.068	10.528	0.000	0.830	0.689
	Accepting change	0.934	0.086	10.819	0.000	0.930	0.865

Note: ¹ SMC = Squared Multiple Correlations, ² CR = Composite reliability, ³ AVE = Average Variance Extracted.

shows that the Average Variance Extracted (AVE) was greater than 0.507. The standardized factor loadings were bounded between 0.930 and 0.609, and the numerical results showed a high degree of internal consistency in the constructs, with all constructs meeting the criteria recommended by Fornell and Larcker [120].

3.3.2. Discriminant Validity

This study investigated the potential relationship between work resilience at the onset of the epidemic COVID-19, in order to investigate the complex effects of mediating variables [121]. Fornell and Larcker [120] state that the square root of the AVE of each component needs to be greater than the correlation coefficient between the construct and the other constructs. They explain that discriminant validity should also consider the correlation between convergent validity and composition, as the diagonal line in

Table 4. Discriminant validity of the measurement model.

	AVE	Perception of COVID-19	Social Support	Career Resilient
Perception of COVID-19	0.624	0.790
Social support	0.507	0.490	0.712
Career resilient	0.749	0.557	0.744	0.865

Note: The items on the diagonal in bold represent the square roots of the AVE; off-diagonal elements are the correlation estimates.

is the square root of AVE. All of them meet the standard structure with discriminant validity, indicating that this study has discriminant validity. In this study, the VIF between the constructs was less than 5 [122] and the Discriminant Validity HTMT < 0.90 [123] was confirmed to be free of co-linearity problems before model analysis.

3.3.3. Structural Model Analysis

The structural model analysis was performed by the Maximum Likelihood Estimation method, and the indicators of analysis included model fit, significance check of study assumptions, and explainable variation (R2). This study applies Jackson, Gillaspy, and Purc-Stephenson’s [124] study of several SSCI papers as a blueprint for applying model fit analysis and reports the results of this study using the eight most widely used fitness metrics.

Since the sample size of this study was larger than 200, the cardinality was too large, resulting in poor fit, so the fit was corrected by the Bootstrap method [125]. Bollen-Stine Bootstrap corrected for model fit and showed that the model fit met the minimum requirements (χ² = 1202.416, df = 933, p < 0.001, χ²/df = 1.288, RMSEA = 0.028, TLI (NNFI) = 0.978, CFI = 0.979, GFI = 0.914, AGFI = 0.903). The composite reliability (CR) values of the scales were all greater than 0.773 or more, and the average variance extracted (AVE) for all constructs was more than 0.516. This indicates that the results of this study are an acceptable model and that the variables observed in the scales may have a significant effect on the corresponding latent variables.

According to

Table 5. Regression coefficients.

Dependent Variable	Independent Variable	Unstd. 1	S.E. 2	Unstd./S.E.	p-Value	Std. 3	R2 4
Social support (SS)	perception of COVID-19 (COVID-19 PE)	0.390	0.066	5.891	0.000	0.490	0.240
Career resilient (CR)	perception of COVID-19 (COVID-19 PE)	0.244	0.067	3.646	0.000	0.253	0.602
	Social support (SS)	0.752	0.103	7.302	0.000	0.620

Note: ¹ Unstd. = Unstandardized regression coefficients, ² S.E. = Standard Error, ³ Std. = Standardized regression coefficients, ⁴ R² = Explainable variations.

, we can understand the pattern path coefficients of the epidemic in terms of social support and resilience of the industry. The perception of COVID-19 (β = 0.390, p < 0.001) significantly positively affected the social support (SS), family and friends support, state support, and career support. Perception of COVID-19 (COVID-19 PE) (β = 0.244, p < 0.001) significantly affected career resilience (CR) through mediated social support (SS) (β = 0.752, p < 0.001). Self-efficacy, occupational beliefs, and acceptance of changes in the career are important influencing factors in adapting to changes in the career under the spread of the epidemic, and there is a positive relationship with social support. In short, the study results support the model assumptions and are positively correlated. The standardized regression coefficient of perception of COVID-19 (COVID-19 PE) on explaining social support (SS) was 0.490, and the coefficient reached a significant level with an R2 of 0.240, which means that perception of COVID-19 could explain 24% of the explanatory variance of social support; the explanatory power of perception of COVID-19 (COVID-19 PE) on explaining career resilience (CT) mediated by social support (SS) was 60.2%. The explanatory power of perception of COVID-19 (COVID-19 PE) on career resilience (CT) mediated by social support (SS) was 60.2%, as Figure 2 shows. As a result, with the mediation of social support, the self-efficacy and career beliefs of individuals can be enhanced, and the industrial resilience of the tourism industry against the epidemic can be prolonged to adapt to changes in the career or to challenge new jobs.

3.3.4. Mediating Effect Analysis

The mediating effect is the effect of the independent variable to influence the dependent variable through the mediating variable. The mediating variable is a variable that is closer to the outcome variable than the predictive variable, and the mediating variable itself is also a causal (endogenous) variable. The indirect effect of mediating variables includes causal path mediating effect, indirect effect coefficient product, and indirect effect bootstrapping method.

The most commonly used method for detecting mediation effects is Baron and Kenny’s [126] causal path method. The causal path method has been mainly criticized for having too low statistical power [127,128]. The cause–effect approach lacks a quantitative verification process for indirect effects. In addition, the coefficient product method Sobel’s z test [129,130] can also be applied to examine the mediation effect. Although Sobel’s z test is often used, it is mainly used to supplement the causal path method, not to replace it. However, Sobel’s z test assumes that the indirect effect must conform to a normal distribution [131,132], so even if z > 1.96 it does not mean that the indirect effect is necessarily significant.

In some studies related to indirect effects, bootstrapping has been shown to have more statistical power in determining indirect effects than causal and coefficient product methods [133,134]. One of the greatest advantages of the bootstrapping method is that the estimation of the indirect effects does not require a normal distribution sampling assignment of the indirect effects unlike the coefficient product method (e.g., B-K method). The bootstrapping method is a repeat sampling method for the original sample, in which the product of a*b is estimated once when a sample is generated. Hayes [135] recommends that this process be repeated at least 1000 times, with 5000 times being preferred. This study was analyzed with 5000 times. The bootstrapping method can generate reliance intervals with indirect effects of statistical check power, especially Bias corrected bootstrapping (Bias corrected bootstrapping) [134,136].

The

Table 6. Mediation model indirect effect analysis.

Effect	Point Estimate	Product of Coefficients			Bootstrap 1000 Times
					Bias-Corrected 95%
		S.E.	Z-Value	p-Value	Lower Bound	Upper Bound
Total Effect
Perception of COVID-19 → Career resilient	0.537	0.359	1.496	0.135	0.336	1.458
Total Indirect Effect
Perception of COVID-19 → Social support → Career resilient	0.293	0.099	2.955	0.003	0.159	0.559
Direct Effect
Perception of COVID-19 → Career resilient	0.244	0.332	0.736	0.462	0.047	1.081

below shows the indirect effects of the mediation model, and the confidence interval does not contain 0 [0.159 to 0.559] in the indirect effect of perception of COVID-19 → career resilient, indicating that the mediation effect holds.

4. Conclusions and Discussion

The COVID-19 epidemic has ravaged the world and caused high unemployment, making the future of the tourism industry more uncertain. Therefore, fear of COVID-19 will lead to increased job insecurity. Sustainable management of the tourism industry depends on perception of COVID-19 and community support during the epidemic. The relationship between its impact and career resilience must be understood. Based on the above discussion, the main objectives of this study were to explore (1) There is a mediating effect of epidemic perception through social support (government support, family and friends support, and workplace support) on the career resilient of workers in the tourism industry. (2) Through London’s career motivation theory, we propose that career resilient is the key to career motivation, and conduct an empirical assessment of an individual’s ability to resist and adapt to the work environment. (3) Understand the social support variables of mediation during emergencies and whether they can play an important mediating role. This study used linear structural equation modeling (SEM) to examine the effects of perception of COVID-19 and social support on the career resilience of tourism industry practitioners.

4.1. Discussion

This study examines career resilience models of perception of COVID-19 and social support during the outbreak. The study confirmed that all three sub variables of social support factors have a mediating effect. Workplace support is the most important factor, followed by national rescue policy and help from family and friends. The expansion of social support would enhance the career resilient of practitioners in the face of the epidemic. At the same time, positive social support during the epidemic was effective in improving the psychological burden of childbirth for pregnant women. Educational advocacy and social support can reduce maternal stress and depression during the epidemic [42], thus reducing prenatal depression. In the early stages of the COVID-2019 outbreak, as the number of infections increases, social support can be effective in strengthening resilience and helping people to survive in the face of adversity [27]. Furthermore, social support influences the perception of stressful events, which in turn changes the perception of inappropriate behavior [39,137]. In short, the above literature demonstrates the importance of social support during the COVID-19 period to improve negative emotions or strengthen career resilience through difficult situations, which is consistent with the findings of this study.

The findings show that organizational support during COVID-19 had a positive and significant impact on tourism practitioners’ self-efficacy and professional beliefs, which are consistent with many previous studies. The results of the study showed that positive career support during COVID-19 had a positive effect on frontline workers’ physical fatigue and mental job security. Both informal and formal forms of organizational support can be helpful to employees. Support is most beneficial when employees are in high demand during an outbreak. Organizational and supervisory support influence each other, and supervisory support is most effective when the organization fosters a family-friendly culture [138]. In the context of the COVID-19 contagion crisis, organizational leaders are conflicted between managing employee health and well-being on the one hand and maintaining or restoring profitability on the other [139]. Organizational management and response are relatively difficult. Social psychologists have proposed “identity leadership” as a new and potentially effective form of leadership in the COVID-19 pandemic crisis, which involves leaders in organizations with common interests with other people, and the creation of collective social consciousness [140,141,142]. The idea is to provide managers with the ability to manage effectively and flexibly in different capacities and roles. Therefore, effective and flexible management of different corporate positions.

In addition, support from friends and relatives and government subsidies show a low impact. The reason for this may be related to the industrial impact of COVID-19 and the uncertainty of timing. Tourism industry practitioners are unsure about the control and blockage of domestic and international industry exchanges. The continuity and adequacy of government subsidies (expansion of unemployment relief for the tourism industry, tax breaks, low-interest loans, and the use of tickets such as arts and entertainment vouchers, agricultural tourism vouchers, and other subsidies) are also taken into consideration.

Despite the adversity, practitioners with resilient psychological mechanisms will overcome the negative effects of COVID-19 and work to return to normal. From the current social phenomenon of the tourism industry struggling for survival, it seems that the individual’s adjustment to adversity and resilience are important influencing factors in career resilience. The findings show that psychological factors, whether self-efficacy, beliefs, or acceptance of change, have a degree of influence on individual adjustment and resilience. This has the same results as Pathak and Joshi’s [94] study, which confirmed that psychological capital of optimism, self-efficacy, resilience, hope, and organizational resilience produced important psychological competencies during COVID-19. The positive thoughts of frontline employees during this COVID-19 period mitigated the effects of the fear of change and job insecurity of the impact of the restaurant industry. The fear of COVID-19 has a direct and indirect effect on resilience and subjective well-being and finding ways to cope with adversity and enhance their mental health [143].

The findings of this study suggest that social support played an important role in resilience during the COVID-19 pandemic. The positive self-beliefs and career resilience and acceptance of change were important in generating resilience during the pandemic, in line with studies of career insecurity associated with the pandemic [19], and confirm the importance of positive and strong psychological qualities during the pandemic. Given the devastation and stress caused by a pandemic, personal resilience, and recovery can be strengthened by organizing social support from supervisors and colleagues as well as personal resources and facilities for self-efficacy to survive the impact of the pandemic. Future studies will extend the theoretical application to psychological capital, life satisfaction, or fear of epidemics, or different occupational categories or levels of epidemic response.

The future changes in tourism behavior in response to the epidemic have become a test of career resilience and a focus of future research for tourism industry practitioners. It is important to discover the needs and patterns of mass travel, to understand tourists’ changing habits and behaviors, and to change the established business models to transform and diversify innovations or to prepare for travel security and emergencies [144]. In addition, we should reconsider the growth trend of tourism and accelerate the transformation of sustainable tourism [10]. The spatial deconstruction and reorganization of the post-epidemic landscape have created a diverse tourism product, and the recovery of tourism from the consequences of the COVID-19 health crisis is unforeseen under the indefinite spread of the epidemic, depending in part on the recovery of the global economy [145]. In the travel, restaurant, and air transportation industries, tourists value safety and hygiene, and spatial mobility in their choice of accommodation or transportation [15,146]. The only way to strengthen the industry’s resilience is to effectively master the development of spatial mobility and decentralized activity experiences. Only by reacting quickly and timely and proposing response strategies, making organizational adjustments, and implementing innovative response strategies can organizations effectively integrate resources and transcend changes or transformations during a crisis.

There has been previous research on social support in the general competitive environment, but during the unprecedented COVID-19 pandemic, little has been done to understand the important support factors corresponding to social support for the tourism industry. This study may be an empirical study of the tourism industry to test the impact and importance of social support to the industry during a pandemic. The results may contribute significantly to the tourism literature. In addition, this study can help us understand the characteristics of practitioners’ personal beliefs and effectiveness in career resilience. The results of this study can provide insights into the impact and exploration of crisis events on social support practices, reduce unnecessary waste of resources, and identify key influencing factors.

4.2. Conclusions and Suggestions

Empirical studies have demonstrated the important role of social support in resilience during the COVID-19 pandemic. Positive self-beliefs and career resilience, acceptance of change, and resilience during the pandemic also confirm the importance of positive and strong psychological qualities during the pandemic. Given the devastation and stress caused by a pandemic, personal resilience and recovery can be strengthened by organizing social support from supervisors and colleagues as well as personal resources and facilities for self-efficacy to survive the impact of the pandemic.

Considering the spread of the epidemic, the professional beliefs and adaptations of tourism industry practitioners and positive psychological qualities (self-efficacy, beliefs, and acceptance of change) are important key factors that can help them quickly regain confidence in the industry as the tourism industry struggles to survive. Moreover, a positive organizational culture will help create an atmosphere of trust, which in turn leads to innovation and creativity in the industry. Numerous studies have confirmed the important role of organizational resilience and flexibility in crisis and post-disaster adaptation [94,147,148,149]. In addition, this study recommends that executives should establish resilience and crisis management implementation points to immediately identify priorities at each stage of a crisis to improve employee and organizational resilience as an effective implementation strategy in response to an epidemic crisis [150,151].

Crisis management processes and plans should be developed before a crisis occurs, and important internal management factors, such as organizational vision, organizational culture, human resources, and financial support. It is also necessary to achieve a balanced control point. Furthermore, we recommend establishing a friendly work environment and organizational climate and interpersonal network within the organization to understand, assess, and judge the actions taken by junior employees to respond to crisis and accept change.

Government relief measures and subsidies are relatively important [73,152] to mitigate the impact of tourism and help the tourism industry to weather the storm. This will prevent tourism-related industries (tourism, hotels, restaurants, and air transportation) from experiencing capital difficulties and facing prolonged unpaid leave, layoffs, or closures, and prepare for the next cycle of economic recovery. Government policies can help make the industrial economy more flexible and sustainable by providing different rates of “COVID-recovery tax” [15] year by year. Or, the government can provide priority subsidies to small and medium-sized enterprises in financial distress to assist in restructuring or restructuring. The other way is to establish a two-way communication platform to understand industrial distress and coordinate and integrate industrial resources for a gradual recovery.

4.3. Research Limitations and Future Direction

COVID-19’s extended firing allows people to feel the separation of space transfer and boundary. Future tourism, leisure, and hospitality-related research could focus on value selection in the provision of lodging services, changing consumer behavior patterns, the influence of social media, and the drivers of health and safety travel. These are all highly influential in how travelers evaluate and choose prices in the future. Only by effectively mastering the development of spatial mobility and de-territorialized activity experience, the tourism industry can effectively integrate resources and transcend changes or transformations during the crisis as a strategy to strengthen organizational flexibility and post-disaster changes in the industry.