How Does COVID-19 Risk Perception Affect Wellness Tourist Intention: Findings on Chinese Generation Z

Abstract

Understanding the influencing mechanism of the COVID-19 pandemic on the public’s travel intentions is key to creating effective strategies to restore and enhance confidence in tourism. Based on the theory of planned behavior (TPB), an extended model was proposed to investigate the Chinese Generation Z’s perception of risk and its effect on the consequences of behavioral process toward wellness tourism. A total of 727 respondents were surveyed by using an online questionnaire. The relationships among the perceptions of risks, three key explanatory variables (attitude, subjective norm, and perceived behavioral control) in the TPB, and wellness tourism intention were evaluated. This study verified that both the perceived health risk and the perceived psychological risk negatively impacted the wellness tourism intentions of Gen Z. Attitude is a partial mediator between subjective norms, perceived behavior control, and travel intentions, respectively. The findings are discussed from the perspective of the theoretical and managerial implications, as well as of future research directions.

1. Introduction

According to the statistics of the World Health Organization (WHO), as of October 2022, the COVID-19 pandemic has led to more than 600 million confirmed cases and more than 6 million additional deaths worldwide [1]. Since the outbreak, the continuous spread of the pandemic caused all countries to adopt regional lockdowns, restrictions on entry and exit, and a reduction in commercial mobility and social activities. There is no doubt that tourism has been the sector most affected by the COVID-19 pandemic. According to the statistics of the United Nations World Tourism Organization (UNWTO), the number of global inbound/outbound tourists decreased from 1.4 billion in 2019 to barely more than 0.4 billion in 2020 and 2021 [2]. On the one hand, people are inclined to have the desire to travel after staying at home for a long time. On the other hand, the COVID-19 pandemic-related risk is the main concern that affects people’s responses to traveling activities [3]. People may involve health-related risks in their whole process of traveling. Therefore, understanding the decision-making processes and intentions of tourists’ participation in tourism is significant for the recovery of tourist destinations and the tourism sector [4]. Academic research on the relationship between the COVID-19 pandemic and the tourism sector is increasing [5,6], with more focus on the negative impact of the pandemic on the development of tourism [7,8,9,10]. A few studies have focused on tourists’ behavioral intentions in response to the pandemic [3,5,11]. On the other hand, demographic change today creates both new challenges and new opportunities for tourism development [12]. The studies of tourists’ behaviors increasingly focus on specific population groups, such as Generation Z [13,14,15]. However, the research on the impact of the pandemic on tourism activities from the perspective of these young people is inadequate. In other words, their intentions to engage in travel activities are still unclear in the context of the COVID-19 pandemic.

Generation Z (hereafter Gen Z) refers to the generation born between 1995 and 2012, corresponding to the current age range of 10~27 [16]. In the context of Western society, Gen Z comes after the previous Gen X (1965–1980) and Gen Y (1981–1994). In Chinese society, the applicable intergenerational labels are “post-95” and “post-00”. Figure 1 shows the market segmentation of China’s generation associated with travel participation, which was derived from the statistics of Chinese population distribution [17,18]. Experts forecast that this large population could contribute 40% of the entire country’s consuming power [19]. Meanwhile, having diverse options, being experience-driven, engaging in online review-making, etc. are considered the main characteristics of Gen Z travelers [20]. For example, a recent report from the Chinese International Travel Monitor (CITM) indicated that post-90s travelers devoted 36 percent of their income to travel, with an average of four trips a year [21]. Even though the COVID-19 pandemic is still not over, tourism businesses are still expecting the relaxation of restrictions. Especially those major travel groups, like online travel agencies, have high hopes riding on Gen Z.

Strauss and Howe’s intergenerational theory states that a generation is developed by the people born at a certain life stage who have a relatively similar position in history, such that they have a common collective role [22]. People of the same generation usually experience similar political, economic, and social events and life experiences, thus forming similar ideologies, values, and worldviews. As Gen Z has gradually stepped into society, the influence of this group of young people, known as “igenerations”, “post-millennials”, or “zoomers”, is also enlarging. The birth and growth of this group are occurring at a rapid stage of China’s economic development, and the children of China’s main middle-class and affluent-class families are also included in this group. Therefore, the young consumption power dominated by Gen Z is gradually becoming the backbone of China’s consumer market. Academic institutions and market agencies are paying increasingly more attention to the consumption concept, behavioral characteristics, and lifestyle of Gen Z. Particularly, in terms of tourist consumption, researchers found that Gen Z is different from other intergenerational groups in terms of the characteristics of their consumption [23,24], destination preference [12], word-of-mouth evaluation [25], etc., and they are more personalized [26], focusing on the emotional experience brought about by consumption [27]. Their tourism consumption and decision-making have a very significant impact on today’s tourism market, particularly on destination marketing and strategies [13,28].

As a tourism mode beneficial for physical and mental health, wellness tourism is gradually becoming popular, and the Chinese tourism service supply related to the health sector is also increasing [29,30,31,32]. The COVID-19 pandemic has aroused people’s concerns about their health. For example, Yang et al. found that more than half of the respondents in their study had taken the initiative to exercise during the quarantine for COVID-19 prevention [33]. In addition, people staying at home for a long time began to pay increasingly more attention to their mental health [34]. For example, the research of Lu et al. found people had a compensatory intention for outdoor travel to alleviate the negative psychological and emotional problems that accumulated because of being at home for a long time [35]. Therefore, the positive effects of wellness tourism against the background of the COVID-19 pandemic deserve more attention. However, the previous studies merely focused on the negative impact of the pandemic itself, instead of paying adequate attention to the research on the intentions of individuals to participate in wellness travel under the influence of the pandemic. In recent years, wellness tourism has been gradually becoming popular in China [36]. It is integrating with the developing personal care, culture, and entertainment [37,38]. Some Gen Z tourists who care about their physical wellness and entertainment have been attracted by wellness tourism. However, there are relatively few studies on the participation of Gen Z in the domain of wellness tourism at present. Therefore, studying the intention of Gen Z to participate in wellness travel against the background of the COVID-19 pandemic will help to enrich the research on the tourism consumption of Gen Z and develop the market segment of wellness tourism [39].

This study contributes to the current knowledge of Gen Z wellness tourists’ behaviors and the expanded TPB model in several ways. First, it subdivides perceived risks and examines their effects on Gen Z’s behavioral intentions toward wellness traveling in the context of the COVID-19 pandemic. Second, this study investigates the relationships between the perceptions of risks and each key variable of TPB (i.e., attitude, subjective norm, and perceived behavioral control) in the expanded constructs. Third, it assesses the potential mediating role of attitude between the perceived risks (i.e., health risk and psychological risk), the other two key variables (i.e., subjective norm and perceived behavioral control), and travel intention. This study is another attempt to understand the consumer behavior of young-generation tourists toward wellness tourism in the context of the COVID-19 pandemic.

2. Theoretical Basis and Research Hypotheses

2.1. Perceived Risk of Tourism

The definition of perceived risk mainly originated from the subjective perception of risk. In the field of consumption, it refers to consumers’ views on the overall negative effects of an action plan based on their assessment of possible negative results and the possibility of such results [40]. In the tourism domain, perceived risk is usually defined as the perception of negative events below an acceptable level, which may occur not only during travel but also in the generation stage of travel intention [38]. Conceptually, perceived risk is heavily correlated with perceived uncertainty [41]. The main feature of perceived risk can help tourists predict and make decisions to avoid going to certain places. Previous studies confirmed that the inducing factors of such uncertainty were that travelers or customers perceived their physical health to be at risk, e.g., from political situations, war, pandemic, crime, nature disasters, and terrorism [42,43]. Roehl and Fesenmaier divided risks into three types, namely destination risk, physical-equipment risk, and vacation risk [44]. Reisinger and Mavondo categorized travel risk as terrorism risk, health financial risk, and socio-cultural risk [42]. Regarding the perception of a specific risk such as the pandemic, researchers agreed to emphasize that physical health risk turned out to be the primary and main risk perceived by the majority of visitors [4].

Along with the health risks caused by a series of pandemics, such as SARS, MERS, and Ebola, a growing amount of research focuses on the risk to understand tourist’s perceptions [45,46,47,48,49]. In the early outbreak stage of the COVID-19 pandemic, people had perceived a high degree of health risk when traveling to a destination. With the vaccination and development of specific drugs, the infected population and the case fatality rate have shown a significant downward trend. People’s perceptions of the health risks (e.g., pneumonia and respiratory breakdown) caused by the virus have changed with the shift in the pandemic [50]. However, as the virus becomes more mutable and transmissible, the health risk caused by viral infection is still present [51,52].

The WHO classified the risk of travel, including the mental health risk, as an environmental risk [53]. Roehl and Fesenmaier concluded different risk dimensions by studying the factors of psychological risk related to travel, such as vacation risk and destination risk [44], reflecting tourists’ perceived risks in travel and influencing their negative evaluation of satisfaction and image of the destination. Chew and Jahari summarized such risk dimensions as social–psychological risks [54]. In addition to health risks, previous tourism and hospitality studies classified several risks factors: functional risk (physical and equipment problems), crisis risk (natural disasters or terrorist attacks) and political risk (being caught by police) [55]. In terms of tourism destination community, there have been a series of efficient implementations of pandemic control policies that prevent COVID-19 from spreading in China. There are certain limits to movement imposed on visitors or consumers at travel destinations. With the development of epidemiological investigation and tracking technology, such perceived uncertainty is inevitably involved when people have contact with potential patients. After acquiring such information and knowledge associated with the COVID-19 pandemic, people may have concerns that include fear, unnecessary tension, anxiety, and discomfort related to traveling [42,56]. Psychological risk captures people’s mental concerns caused by the uncertainty of traveling during the pandemic. As such, this study considered psychological risk as another uncertainty that tourists may face during travel.

As Gen Z has gradually become the mainstream consumer group in tourism and hospitality [57,58], researchers began to pay attention to the characteristics and influencing factors of the decision-making behavior of Gen Z [15,59,60]. Greatly affected by technological progress and social media, Gen Z is more dependent on the internet when shopping [28], more susceptible to the impact of social media in the selection of tourism destinations [61], and more favorable toward energy-saving and environment-friendly travel styles [62]. As young people are mainly seekers of sensation and experience, they are more likely to be involved in adventures during travel [63]. Williams and Baláž found that compared with older people, young people tend to be more risk tolerant [64]. Qi et al. investigated US college students’ perceptions of risk associated with traveling to China as the Olympic Games host country; they found that younger people did not appear to be homogeneous in the perceptions of risk [65]. Golets et al.’s research found that younger tourists did not reflect a relatively significant propensity for risk taking [4]. However, there have been few studies on tourist decision-making behaviors and the perceptions of tourism risks of Gen Z. As mentioned, this study incorporates both perceived health risk (PHR) and perceived psychological risk (PPR) as factors that affect Generation Z tourists’ perceptions of risk during the COVID-19 pandemic. PHR refers to the perceived risk of infection with SARS-CoV-2 under the influence of the pandemic [66]. PPR refers to the perceived risk of individuals measuring travel uncertainty associated with mental issues (anxiety, stress, or fear) while participating in wellness travel [67,68].

2.2. Theory of Planned Behavior (TPB)

TPB is one of the most widely used social psychological theories that aims to predict human decision-making and behavior [69]. Its applicability to and predictive power over a variety of human behaviors has been demonstrated through many empirical studies [70]. In the field of tourism research, TPB is commonly applied to explain the decision-making processes and behaviors of travelers [71,72]. TPB is an extension of the theory of reasoned action (TRA), which is the original theory that predicts an individual’s behavior. Later, TRA was extended to the current TPB to overcome this constraint of willpower to create a new conceptual framework for predicting and interpreting human behavior [73]. TPB involves determining an individual’s behavioral intention using attitude, subjective norms, and perceived behavioral control [74,75]. The attitude toward behavior is the first salient component of tourists’ intentions [76]. The concept of attitude refers to an individual’s behavioral beliefs and general feelings regarding a positively or negatively described behavior [60]. The subjective norm is another crucial predictor of tourists’ travel intentions [77]. It refers to an individual’s view of performing a particular behavior or not while feeling social pressure [78]. The perceived behavioral control is also a key determinant of travelers’ behavior intentions. Perceived behavioral control is known as a “control belief” and refers to how an individual perceives the difficulty of completing a particular action [79]. Travel intention refers to an indication of an individual’s readiness to perform behavior to engage in traveling. It reflects the consequences of a mental process that results in an action into behavior [80].

TPB has been extensively used in social psychology and management, and researchers in the tourism field also use TPB to help with interpreting tourists’ behavioral intentions. For example, the consumption scenario is applied to reception services, such as hotels [81] and catering [82]. To highlight its role in interpreting and predicting the behavioral characteristics of tourists or dwellers, studies predicted future survey results by exploring the participants’ opinions on tourism development [83]. Goh and Baum employed the TPB framework to understand Generation Z staffs’ perceptions toward their working in quarantine hotels. They found that young employees treated such an opportunity as meaningful while facing the risk [84]. Juschten applied the extended TPB model to predict tourists’ behaviors and activities in the context of climate change [85]. The study of the TPB model was applied to various tourism forms to interpret tourists’ behaviors and activities, such as cultural heritage tours [86], creative tours [87], and wine tours. Lee et al. used TPB to interpret the intentions of Korean wine tourists to participate in tour groups [88], while Zhang et al. integrated TPB into a framework for resolving conflict behaviors to study the practical problems of cultural heritage tours [89]. Meng and Han used TPB to study the decision-making processes of working vacationists [90]. Some studies focused on the behavioral characteristics of tourists’ travel time [91]. The application of the TPB model is becoming increasingly more extensive. However, in terms of the forms of tourism, research on wellness tourism is still rare. Only Hudson introduced life pressure and health and tried to explore the behavioral intention of wellness tourists in the United States [92]. Additionally, for intergenerational tourists’ behaviors and activities, while there is still little research on the interpretation and prediction of tourists’ behaviors and activities among intergenerational groups, some scholars considered the service staff [93]. Therefore, studying the intention of Gen Z to participate in wellness travel can enhance and enrich the theoretical interpretation of this phenomenon and help the sector by providing a strategic reference for the expansion of potential markets. Based on the TPB theoretical model, this study considered the attitude, subjective norm (SN), and perceived behavioral control (PBC) as three antecedent variables and the intentions of tourist behavior as the research variable in the conceptual framework.

2.3. Perceived Risk and TPB

Kozak et al. pointed out that the perception of risk is a core factor in the decision-making processes of tourists, and it may even change the rational decision-making of travel or destination selection [94]. Thus, the combination of perceived risk and TPB is a popular way to study the perceived risks of tourism [95]. Since the last SARS epidemic, China has not experienced a public health crisis in nearly two decades. Most young people grew up in a relatively safe and harmonious social environment [96]. Therefore, it is necessary to understand such generational groups’ perceptions of risk and their responses regarding travel activities. Scholars considered that the higher the level of risk perceived by tourists, the more likely it is for them to have a more negative attitude toward tourists’ behaviors and activities [97]. Recent studies demonstrated that the perceived risks involving the pandemic have a generally negative impact on people’s attitudes toward traveling, whereas such relationships are not further assessed on the specific demographical groups [98]. PBC reflects the judgment of an individual’s experience, ability, resources, and other capacities in participating in tourists’ behaviors and activities. As the current pandemic-control regulations are irresistible for individual tourists, the perceived pandemic risk is higher, the resistance against individual behavior is stronger, and the PBC is more negative [78]. Based on the above view, the following hypotheses were established:

Hypothesis 1 

(H1).PHR will negatively affect Gen Z’s attitude toward participation in wellness tourism.

Hypothesis 2 

(H2).PPR will negatively affect Gen Z’s attitude toward participation in wellness tourism.

Hypothesis 3 

(H3).PHR will negatively affect Gen Z’s PBC toward participation in wellness tourism.

Hypothesis 4 

(H4).PPR will negatively affect Gen Z’s PBC toward participation in wellness tourism.

Extensive studies using the TPB model support the positive impact of three antecedent variables on behavioral intention [72]. For example, Villace indicated that attitude, SN, and PBC had a positive impact on the intention to participate in vacation tourism in a crisis scenario, while the perception of health risks negatively impacted the intention to participate in vacation tourism [99]. Previous research has shown that a memorable historical event that occurs during a person’s “adulthood” shapes the long-term core values that influence a person’s life, preferences, attitudes, and behaviors [100]. Especially with the younger generations, such an unprecedented epidemic in public health might deeply affect their perceptions and expectations of the future [101]. In regard to this, it is reasonable to evaluate their attitudes, SN, and PBC and their influence on travel intention toward wellness tourism. Juan studied the behavioral intentions of Korean tourists to participate in inter-Korean border tours and pointed out that attitude, SN, and PBC had a positive impact on travel intention (TI), but the perception of war risk negatively impacted such an intention [102]. Therefore, the following hypotheses were established:

Hypothesis 5 

(H5).Attitude will positively affect Gen Z’s TI toward participation in wellness tourism.

Hypothesis 6 

(H6).SN will positively affect Gen Z’s TI toward participation in wellness tourism.

Hypothesis 7 

(H7).PBC will positively affect Gen Z’s TI toward participation in wellness tourism.

Hypothesis 8 

(H8).PHR will negatively affect Gen Z’s TI toward participation in wellness tourism.

Hypothesis 9 

(H9).PPR will negatively affect Gen Z’s TI toward participation in wellness tourism.

2.4. Attitude as a Mediator

Undeniably, attitude is a prominent factor that influences an individual’s behaviors in decision-making processes [103]. In this study, another purpose is to understand the role of attitude toward wellness tourism. In the original proposed TPB model are three independent antecedents that contributed to behavioral intention; however, scholars found that there exists internal causal paths among the constructs. For instance, scholars indicated that SN has a positive impact on attitude, which showed a strong explanatory power of behavioral intention [104,105,106]. The positive impact of PBC and attitude was also demonstrated by researchers [79]. In the context of preventing viral infection, an individual’s SN and PBC have increasingly significant effects on behavior in terms of whether to stay away from risk [67]. In contrast, Xu et al. assessed Hong Kong residents’ perceptions of and intentions to engage in local tourism. He found that PBC has no significant influence on attitude whether during or after the pandemic [68]. Furthermore, Yang and Lee found PBC also has no significant influence on behavioral intention in terms of accommodation-sharing services [107]. Wang et al. asserted that the context of behavioral constraint regarding the pandemic has created a situation for assessing the mediating role of attitudes in TPB [108]. For example, Bae and Chang found that, in Korea, people are knowledgeable about pandemic-preventing measures, including movement restrictions and social distancing. They may face social pressure to comply with preventive behaviors, whereas such an SN factor used to be the major driving factor of travel intention [109]. Therefore, it is reasonable to predict that the young generation’s beliefs toward and evaluations of participating in travel will be affected by such social pressures and behavioral constraints. As such, this study will assess the mediating role of attitude in TPB, and the following hypotheses were established:

Hypothesis 10 

(H10).SN will positively affect Gen Z’s attitude toward participation in wellness tourism.

Hypothesis 11 

(H11).PBC will positively affect Gen Z’s attitude toward participation in wellness tourism.

Hypothesis 12 

(H12). Attitude will mediate the effect of Gen Z’s SN on TI toward participation in wellness tourism.

Hypothesis 13 

(H13). Attitude will mediate the effect of Gen Z’s PBC on TI toward participation in wellness tourism.

Regarding the relationships among perceptions of risk, attitudes, and travel intentions, attitude is always a critical precursor of travel intention [110]. Concerning the reality of the context, the main consideration before participating or thinking about participating in tourism activities is the perceived risk of the pandemic. Unlike previous studies that treated risk factors as interference in the process of behavioral intention, this research has placed risk perception as the primary influencing factor [67,107]. For example, Bae and Chang found that attitude exhibited a mediating effect between affective risk perception and behavioral intention [109]. Lee demonstrated that attitude acts as a mediator between five types of risks and the intention to use online banks [111]. In this study, we postulate attitude will mediate the effect between both the perceived health risk and the perceived psychological risk and the travel intention toward wellness tourism, respectively.

Furthermore, the following hypotheses were established:

Hypothesis 14 

(H14). Attitude will mediate the effect of Gen Z’s PHR on TI toward participation in wellness tourism.

Hypothesis 15 

(H15). Attitude will mediate the effect of Gen Z’s PPR on TI toward participation in wellness tourism.

The specific structure of the proposed research model presented below illustrates our hypotheses (Figure 2).

3. Design of Research

3.1. Measurement of Variables and Sources of Questionnaire

After reviewing the literature, the comprehensive questionnaires used in previous studies were applied in this study. Before designing the questionnaire and considering that the sample collection objects were in mainland China, to ensure the consistency and semantic accessibility of the items measured by the scales, this study adopted the method of “translation and back-translation”. First, a professional translator was invited to translate the items of the questionnaire with the authors of this paper, and then another professional translator and a professional tourism scholar with higher English proficiency were invited to back-translate the text. To ensure the translation accuracy and professionalism of the items in the questionnaire, the authors discussed and revised the translated and back-translated texts together with the scholars involved in the translation and determined the first draft of the questionnaire.

The questionnaire for PHR and PPR borrowed ideas from several past research projects on travel during the COVID-19 pandemic [68,107,109] and from Chew and Jahari’s [30] perceived risk of tourism in Japan. The measurements of attitude, SN, PBC, and TI in the TPB model were based on the research questionnaires of Juan et al. and Liu et al. [34,37,60]. The questionnaire designed in this study was mainly divided into two parts. The first part was about perceived risks, attitude, SN, PBC, and TI, and the second part was about the demographic profile of the respondents. A seven-item Likert scale (1—extremely disagree; 7—extremely agree) was adopted to grade all the items in the questionnaire. The respondents were required to grade each item.

3.2. Data Collection

Conceptually, the target respondents of this study should meet the criteria of the age range from 10 to 27 years old. However, most adolescents are not allowed to travel privately without their legal guardians (parents) accompanying them. Moreover, some young people (ages 15–17) are in high school and under pressure to enter college/university. Therefore, the optimal samples of this study are limited to the age range of 18–27 years old. To ensure the reality and content validity of the responses to the questionnaire, respondents consisted of college students, graduate students, and those who had graduated from university and were employed but were younger than 27 years old. An electronic version of the questionnaire was prepared and distributed on the online platforms Sojump.com and Credamo.com. The samples were collected from the middle of June 2022 to the end of July 2022. A total of 794 copies of the questionnaire were distributed and returned. The returned responses to the questionnaire were screened. A total of 67 copies were considered invalid, and there were 727 valid questionnaires collected, with an effective response rate of 91.6%. To understand the background information of each respondent, descriptive statistical analysis was implemented. To test the goodness of fit of the model and the proposed hypotheses, Amos 22.0 was used for the confirmatory factor analysis (CFA) and structural equation modeling (SEM).

The sample statistics showed that there were 332 female respondents, accounting for 54.3% of the total, compared with 45.7% male respondents. In terms of education level, most respondents were highly educated. Most of the educational backgrounds were university and college level, and the number of those with higher education is increasing. Among others, 31.4% were undergraduates, and 142 had a master’s degree or above, accounting for 19.5%. This study grouped the participants in 2-year increments for the range of ages from 18 to 27. The data showed that the age distribution of the respondents was relatively balanced, which indicated that the samples collected covered all age groups of Gen Z. In terms of marriage, the vast majority of respondents were unmarried, while 89 were married, accounting for 12.2%. According to the latest data, the average age when Chinese males and females marry is about 28.67 [112], which indicates that the age of marriage of Gen Z in the sample was also gradually increasing. In terms of monthly income, most of the respondents were in the middle- and low-income groups. The range below RMB 6000 accounted for 77.4%. See

Table 1. Respondents’ demographic profile.

Demographic Profile		Frequency	Percentage
Gender	Male	332	45.7
	Female	395	54.3
Educational level	Secondary school or lower	79	10.9
	High school/technical school	116	16.0
	College	162	22.3
	Bachelor	228	31.4
	Postgraduate or higher	142	19.5
Age	18~19	166	22.8
	20~21	181	24.9
	22~23	152	20.9
	24~25	127	17.5
	26~27	101	13.9
Marriage status	Single	638	87.8
	Married	89	12.2
Monthly income (RMB)	2000 or less	204	28.1
	2001~4000	191	26.3
	4001~6000	167	23.0
	6001~10,000	111	15.3
	10,001~20,000	45	6.2
	20,001 or more	9	1.2

for the specific data.

4. Data Analysis and Findings

Before the model analysis and test, the appropriateness of the data was checked, and the missing values were interpolated with the sequence average. SPSS25.0 was used to test the model and evaluate its normal distribution. According to the testing results, the kurtosis, skewness, and absolute value of each item met the criteria, indicating that it met the criteria for a normal distribution. Then the validity of the model was assessed to reduce the errors by the common method variance resulting from the use of simultaneous data in analysis. Based on Harman’s single factor test, all constructs were composed to one construct in factor analysis. All variables accumulated as 47.26% (<50%), for which no common method variance error exists [113,114]. Amos 22.0 was used in the test of the measurement model for confirmatory factor analysis. According to Fornell and Larcker, Cronbach’s α values above 0.7 represent good reliability [115]. Cronbach’s α of each variable in this study was higher than 0.7, representing the good reliability of the measurements. CFA was used to test whether the indicators of each measurement model conformed to the theoretical model. The aggregation validity test is mainly based on two indicators: the composite reliability (CR) and average variance extracted (AVE). The analysis results showed that the CR value of each variable was higher than 0.8, and the AVE was higher than 0.5, indicating that the measurement model had good aggregation validity. See

Table 2. Results for the CFA.

Items	Std FL	SMC	α	CR	AVE
Perceived health risk			0.892	0.940	0.759
PHR_1 There is a likelihood of me contracting COVID-19 while traveling.	0.876	0.767
PHR_2 I am concerned that COVID-19 caused serious issues to my body.	0.826	0.682
PHR_3 Compared with other diseases; the risk of COVID-19 spread is higher.	0.924	0.854
PHR_4 I am worried that I will come into contact with the COVID-19	0.808	0.653
PHR_5 I am worried about my family members contracting COVID-19	0.915	0.837
Perceived psychological risk			0.864	0.827	0.614
PPR_1 I am worried that my wellness traveling will not be compatible with my expectation	0.768	0.590
PPR_2 I am worried that something uncertain would happen and change my travel plan.	0.802	0.643
PPR_3 I am worried that my wellness traveling will make me feel psychologically uncomfortable.	0.781	0.610
Attitude toward wellness tourism			0.927	0.880	0.647
ATT_1 I believe my wellness traveling is valuable.	0.754	0.569
ATT_2 I believe my wellness traveling is beneficial.	0.762	0.581
ATT_3 I believe I would enjoy my wellness traveling.	0.862	0.743
ATT_4 I believed I would be satisfied with my wellness traveling	0.834	0.696
Subjective norms			0.889	0.909	0.673
SN_1 My family would support me to travel.	0.857	0.734
SN_2 My friends would support me to travel.	0.695	0.483
SN_3 People who are important to me think I should go to travel for wellness.	0.958	0.918
SN_4 People around me would support me to travel.	0.901	0.812
SN_5 People who are close to me understand that I engage in wellness tourism	0.646	0.417
Perceived behavioral control			0.943	0.912	0.722
PBC_1 I am capable of engaging in wellness tourism	0.846	0.716
PBC_2 I have sufficient resources, time and opportunities of engaging wellness tourism.	0.912	0.832
PBC_3 I am confident if I want to, I can engage in wellness tourism.	0.776	0.602
PBC_4 Engaging in wellness tourism is not difficult for me.	0.859	0.738
Travel Intention			0.857	0.939	0.793
TI_1 I would like to engage in wellness tourism someday.	0.884	0.781
TI_2 I believe it’s time to alleviate the travel restrictions.	0.906	0.821
TI_3 I believe safety measures can allow me to travel.	0.876	0.767
TI_4 I have confidence to travel during pandemic.	0.895	0.801

for the specific data. Discriminant validity refers to the degree of differentiation of the potential variables. According to the criterion formulated by Fornell and Larcker [115], as shown in

Table 3. Results for the discriminant validity.

	PHR	PPR	ATT	SN	PBC	TI
PHR	0.871
PPR	0.476	0.784
ATT	−0.428	−0.138	0.804
SN	−0.339	−0.267	0.459	0.820
PBC	−0.176	−0.562	0.368	0.319	0.850
TI	−0.217	−0.364	0.485	0.426	0.372	0.891

, the square root of the average variation extraction was greater than the diagonal value in the column and row, which meant that the self-correlation was greater than the correlation with other structures, and thus, the measurement model structure had good discrimination validity.

In the measurement test, the reliability and validity of the model were both in line with the recommendations of scholars. Therefore, the model had favorable adaptability. In the structural test of the model, the structural validity was tested by measuring the adaptability of the structural model and other indicators. The maximum likelihood method was used to estimate the parameters, and it was found that the overall structural model had good fitness and all the indicators were within a reasonable range (shown in

Table 4. Results of model fit indices.

	Χ2	df	Χ2/df	Probability Level	RMSEA	GFI	CFI	IFI	TLI
Computed value	127.621	68	1.877	0.001	0.048	0.921	0.944	0.974	0.937
Recommended value			1–3		<0.05	>0.9	>0.9	>0.9	>0.9

) [116,117].

A total of 11 hypotheses in the structural model with an impact on the path were tested, as shown in

Table 5. Structural model test results and hypothesis test outcomes.

Hypotheses	Hypothesized Path	Path Coefficient	S.E.	p	Remarks
H1	PHR → ATT	−0.201	0.136	**	Supported
H2	PPR → ATT	−0.031	0.076	0.683	Not supported
H3	PHR → PBC	−0.027	0.061	0.107	Not supported
H4	PPR → PBC	−0.208	0.07	**	Supported
H5	ATT → TI	0.338	0.092	***	Supported
H6	SN → TI	0.218	0.075	**	Supported
H7	PBC → TI	0.253	0.136	**	Supported
H8	PHR → TI	−0.127	0.076	*	Supported
H9	PPR → TI	−0.359	0.048	***	Supported
H10	SN → ATT	0.273	0.095	***	Supported
H11	PBC→ATT	0.124	0.057	**	Supported

*: significant at the 0.05 level; **: significant at the 0.01 level; ***: significant at the 0.001 level.

and Figure 3. First, PHR had a significant negative impact on attitude (β = −0.201, p < 0.01) and TI (β = −0.127, p < 0.05), indicating that the higher the respondents’ perception of physical risk associated with COVID-19 infection, the more negative their attitudes and TI. Therefore, H1 and H8 were well-supported. However, the negative effect of PHR on PBC was not significant. It reflected that Gen Z’s perceived risk regarding health would not influence their perception of behavioral control. Thus, H3 was rejected. Second, PPR had a significant negative impact on PBC (β = −0.208, p < 0.01) and TI (β = −0.359, p < 0.001), indicating that the higher the perception of risk imposed by pandemic control on tourism, the lower the PBC and the lower the TI. However, the negative impacts of PPR on attitude (β = −0.031, p = 0.683) was not significant, indicating that Gen Z perceived that the risk of the pandemic on travel was not significantly related to their attitude. Their perceptions and beliefs toward engaging in wellness tourism would not be affected by their perceived risks regarding social context. Therefore, H4 and H9 were well-supported, but H2 was not. In the original TPB model, attitude, SN, and PBC all had a significant positive impact on TI, indicating that the higher the attitude, SN, and PBC of the respondents, the stronger their TI. Therefore, H5, H6, and H7 were well-supported. Third, the path coefficients showed that both SN and PBC had affected attitude (SN: β = 0.273, p = 0.095; PBC: β = 0.124, p = 0.057). The findings support that Gen Z’s attitudes toward wellness tourism are positively impacted by subjective norms and perceived behavioral control [68].

As the proposed structural model was established, this study adopted the bootstrap method to assess the mediating role of attitude between SN/PBC/PHR/PPR and TI. Two thousand samples and a 95% confidence level were settled in the bootstrap process. According to Baron and Kenny’s method and criteria, three partial mediation effects were observed in such an effect [118]. The results are shown in

Table 6. Mediating test results for attitude.

Hypotheses	Mediating Path	Indirect Effects	Bio-Corrected 95% CI		Mark
			Lower	Upper
H12	SN → ATT → TI	0.060	0.006	0.102	Partial mediation
H13	PBC → ATT → TI	0.042	0.005	0.117	Partial mediation
H14	PHR → ATT → TI	−0.068	−0.209	0.004	No mediation
H15	PPR → ATT → TI	-	-	-	No mediation

. The attitude in TPB partially mediated the effect of SN on TI, and partially mediated the effect of PBC on TI. It is observed that Gen Z’s SN and PBC have both direct and indirect impacts on travel intention. Thus, hypotheses 12 and 13 were partially supported. Regarding the mediating role of attitude, the result indicated that the effect between PHR and TI was not significant. Although Gen Z’s perceived health risk toward wellness tourism was still salient, its indirect impact on behavioral intention, which was through the attitudinal factor, was not significantly observed. Therefore, hypothesis 14 was rejected. In addition, H15 was also rejected because of there being no significant effect of PPR on ATT, which resulted in no mediation effect observed.

5. Discussion and Conclusions

5.1. Discussion

Unlike previous outbreaks of infectious diseases (e.g., SARS and MERS), the COVID-19 pandemic has profoundly affected the tourism and hospitality industry around the world. People have urgent desires to engage in tourism to maintain their normal lifestyles, mental health, and wellbeing [11,35,119]. However, the perception of risk is a primary restricting factor that cannot be ignored. On the one hand, as the characteristics of Generation Z are more diversified and progressively changing, tourism studies on Gen Z are gradually increasing and are worth exploring [28,120]. On the other hand, the Gen Z segment of the wellness tourism market is also expanding rapidly, which changed the people stereotype of mainstream consumer groups in the wellness market [30,121]. A growing number of studies paid attention to the perceptions of risk in traveling during the pandemic, while a limited number of studies focus on such a generation in assessing its intention to participate in wellness tourism. The main purpose of this study was to understand the impact of the COVID-19 pandemic on the intentions of China’s Gen Z population to participate in wellness tourism. An extended TPB model was proposed to examine the relationships between the perceptions of risk (i.e., PHR and PPR) and the key variables of TPB: attitude, SN, PBC, and TI.

The significant finding of this study verified that perceived risk negatively impacted the intentions of Gen Z groups to participate in wellness tourism. In this study, perceived risk was distinguished into two constructs, namely PHR and PPR, and the impacts of risk on TI were tested against the framework of the TPB theoretical model. The results showed that both health and psychological risks negatively impacted wellness travel intentions. The findings are consistent with those of Liu et al. on outbound tourism in the post-pandemic period and of Hamid et al. on behavioral intentions during the pandemic [61,122]. Furthermore, compared with the health risk, psychological risk has a more significant effect on the intention to travel. In other words, the risk of psychological factors acts as a prominent antecedent of Gen Z’s intention to participate in wellness tourism. This finding supported Bae and Chang’s research regarding the impact of Korean residents’ perceptions of affective risk on behavioral intentions [109]. The perceptions of Gen Z of the COVID-19 pandemic came from the acquisition of information concerning COVID-19 and pandemic control. Because of the high pathogenicity and invisible spread of SARS-CoV-2, the public has been compelled to fear the spread of the virus [123,124]. Thus, the present study added a new explanation for future study to comprehensively understand the factors that negatively impact the intention of Gen Z to participate in travel.

Based on TPB, this study developed an integrated model to research the impact of health risk and psychological risk on attitude and PBC. The results show that PHR had a significant negative impact on attitude, whereas PPR’s effect was not significant. The argument that behavioral attitude can be affected by the perception of holistic risk has been demonstrated by previous studies (e.g., [98]). From the theoretical perspective, the finding noted that Gen Z’s general attitude toward wellness tourism during the pandemic can be affected by the specific perception of risk. Furthermore, the PPR has no stronger effect on Gen Z’s attitude regarding wellness tourism. This specific outcome was inconsistent with Han et al.’s findings on US tourists who perceived that high psychological risk would reflect a positive attitude toward safer destinations [67]. Meanwhile, the results showed that Gen Z’s PPR has a significant effect on PBC, but PHR not. PPR reflects a critical role in predicting PBC rather than PHR, which implies that Gen Z may not be confident in dominating their behavior in participating in wellness tourism as they perceive a relatively high psychological risk during the pandemic. This result supports the finding of Xu et al.,’s study but inconsistent with the results of other previous studies [68,97].

Another relatively important finding of this study is the inter-relationships between attitude and other two key antecedents (i.e., SN and PBC) in the proposed construct. According to results, the positive effect of both SN and PBC on attitude was observed. Compared with PBC, it is found that SN was a stronger predictor of attitude. An individual can perceive social support and encouragement from other people in his/her social context, which will facilitate their positive attitudes toward wellness tourism as well [67]. This indicated that Gen Z’s beliefs toward and evaluations of specific types of tourism can be significantly affected by social factors rather than personally controlled factors. This finding is consistent with previous studies, but inconsistent with Xu’s results [68,79,108].

The last finding of this study is the mediating role of attitude in the extended TPB model. It is found that attitude is partially a mediator between SN, PBC, and TI, respectively. This result indicates that not only can both SN and PBC affect travel intentions directly but also facilitate an individual’s attitude and its effect on travel intention indirectly. Zorlu et al.’s research on camping/glamping tourists found that attitude partially mediated the effect between SN and behavioral intention; this is consistent with our result. However, past research including Zorlu et al.’s did not find a significant mediating effect of attitude between the PBC and TI, which is inconsistent with our findings [79,108]. One possible explanation for such findings is the different groups of respondents. Most young Gen Z today who participate in tourism are largely self-organized, which reflects their confidence in behavioral actions and thus facilitates their attitude and its effect on behavioral intention. Above all, these outcomes also imply the complicated mechanism of forming an individual’s attitudes and behaviors [68].

5.2. Theoretical Contributions

Based on the TPB theory, this study proposed and assessed an integrated extended TPB model that presumed Gen Z’s perceived risks (i.e., health risk and psychological risk) as antecedents and travel intention toward wellness tourism as the outcome. This study also assessed the mediating role of attitude between the relationship of perceived risks and other constructs (i.e., SN, PBC) in TPB. The theoretical findings of this study are mainly summarized as follows.

First, this study explored Gen Z’s risk perceptions regarding the pandemic and its impact on the consequent behavior. There has been a growing number of studies paying attention to the perception of risk and behavioral intention in the context of COVID-19 (e.g., [4,66,125]). This study made an attempt to focus on the future mainstream group of consumers and the notable type of tourism associated with health and well-being. It is believed that this study contributed to a better understanding of the younger generation of travelers in the current context.

Second, this study is more empirical research that links the perception of risk with the original TPB model. Unlike previous studies that place the risk factor as interference in the construct, this study argues that risk is a salient and prior consideration in the decision-making process of travel behavior [4,125,126]. Furthermore, it is observed that both health risks and psychological risks have significant negative impact on travel intention toward wellness tourism. The finding reflects the applicability of the proposed expanded TPB model in the contemporary research background.

Third, in this study, the argument that three key antecedents of TPB are not mutually independent in assessing consumer behavior has been demonstrated again [105]. There have been some empirical studies adopting TPB theory in examining peoples’ travel behaviors during the pandemic. However, few researchers investigated the interrelations among attitude, SN, and PBC. This study demonstrated causal relationships between both normative factor and perceived behavioral factor and attitudinal factors, respectively. These findings contributed to the refinement of socio-psychological theory by considering their interrelations among the key variables [67].

5.3. Managerial Implications

The findings of this study can provide some practical references for destination managers or relevant organizations and institutions involved in wellness tourism. Regarding the significant effect of attitude on TI, wellness tourism stakeholders should try to influence Gen Z’s attitude toward TI. For instance, wellness tourism providers can expand marketing promotion of the health and wellness benefits [127]. Especially for some younger but sub-healthy people, this practical measure can enhance Gen Z’s positive attitude toward wellness tourism. The destination tourism officials also can offer more healthy lifestyle tips for the young generations to emphasize the importance of both physical and mental health through internet social media platforms [128].

In addition, the perception of risk still plays a very important role in predicting Gen Z’s behavioral intention to travel. Therefore, it is necessary to reduce and weaken people’s negative assessment of pandemic-related risks by optimizing the management methods and marketing strategies. For example, tourism officials and destination marketers can promote their launch of safety measures and services for potential customers to reduce their concerns during travel [79]. Furthermore, tourism sectors should attempt to enhance Gen Z’s SN and PBC regarding wellness tourism. Considering the partial mediating effect of attitude between SN, PBC, and TI respectively, it is needed to influence SN and PBC to facilitate a positive attitude toward TI. Therefore, wellness providers and marketers can create a more social environment for Gen Z tourists and their families/relatives/friends to participate in wellness tourism together. Tour operators can create promotions for small-size groups with price discounts to attract more young wellness consumers [129]. As such, tourists would enjoy wellness trips with a higher degree of wellbeing and sense of control [109].

5.4. Research Limitations and Future Directions

This study had the following limitations: First, the study did not take some external factors associated with the pandemic restriction measures into account. Since the outbreak of COVID-19, the government has implemented strict pandemic restrictions for controlling the infected cases. Those issued pandemic restrictions may cause many people to postpone or cancel their planned travel. Therefore, people’s perceptions associated with the pandemic measures issued by the government might have a direct effect on their behavior in making decisions about travel. Future research is recommended to consider this as an interference factor in assessing the process of behavioral intention. The outcome of such future research will contribute to a more meaningful reference for managers’ decisions.

Regarding another limitation, the conclusions drawn from this study only reflect the results of a certain cross-sectional time during the pandemic. As the reality background of this empirical study, the negative impact of the pandemic on wellness tourism can be changed in the future. The consequent change in government policies on pandemic prevention could also lead to several changes in people’s perceptions of risk and travel attitudes. In terms of this, discovering the factors that have influenced people’s perceptions of risk and attitudes are crucial and meaningful. The sequential study is recommended to be conducted in assessing Generation Z’s perceptions and attitudinal behaviors toward tourism, which helps us better understand how humans get through each stage of the crisis and prepare well for the next one.