Perceived Risk and Food Tourism: Pursuing Sustainable Food Tourism Experiences

Abstract

While food can function as a component product of tourism, there remains a need for food tourism to become more sustainable. This study sought to discover what risk dimensions tourists perceive in food tourism and to enhance understanding of what actions and coping behaviors tourists take to lower levels of perceived risk in a food tourism setting. Data were collected in March 2023 for two weeks from Chinese tourists who considered traveling to Jeju Island, South Korea. A total of 303 responses were analyzed using exploratory factor analysis, confirmatory factor analysis, and structural equation modeling to test 13 hypotheses. Results identified physical risk, communication risk, and food-related risk as concerns for tourists, and the results confirmed support for 11 of the hypotheses tested. This study has theoretical implications for supplementing insufficient prior research by examining the risk factors perceived by tourists when participating in food tourism. Practical implications from this study include ways to reduce the level of perceived risk dimensions for sustainable food tourism experiences.

1. Introduction

Food helps to promote social relationships among tourists and enables them to experience new cultures [1]. Consequently, food can be one of the most important cultural resources of a country, which can reflect regional and national identity [2,3]. In this process, food enables tourists to have authentic food experiences outside of their usual environments, as local traditions, culture, and history are expressed through food [4,5]. Food plays a vital role in the experiences tourists have while traveling. Therefore, researchers have looked at various aspects of food in tourism and argued that food itself could give a great deal of value to both locals and tourists as it is recognized as being an economically, culturally, and environmentally sustainable tourism product [6,7]. Thus, food has been shown to play a role in adding great value to tourist destinations, including the identity of a destination as well as contributing to economic, cultural, and regional sustainability. Therefore, increased attention has been paid to the role food has as a sustainable resource in the tourism industry [7,8].

Local food and sustainable food tourism play significant roles in regional development. These can contribute to economic development due to the economic prospects for local farmers and producers. Tourists that experience sustainable food tourism can influence the local economy and buy products that are locally made, providing a positive influence for the economy of the food tourism destination [9]. Likewise, sustainable practices can be influenced by this type of food tourism. Some potential results are the reduction of food waste, the promotion of biodiversity, or the development of organic farming. Thus, contributions to environmental sustainability in food tourism areas can be enhanced through these practices, which may lead to engagement in the community as well as through the interactions that tourists have with the local community [10].

Due to the connection between food tourism and sustainability, food tourism has become associated with sustainable development. There are 17 U.N. Sustainable Development Goals that seek to bring about a more resilient and sustainable future. Goal 12 of these development goals is responsible consumption and production. Reducing food waste and food loss is a critical component of this goal, and seeking to move food tourism in this sustainable direction underscores the need to pursue sustainable food tourism experiences. While sustainable development has been a general concept that has been defined differently by various researchers, the notion of sustainable economic growth is central to achieving the various development goals that have been put forth to achieve it [11]. Development that considers environmental impacts while recognizing that needs related to general well-being and future use of resources are pertinent for sustainable growth are core elements of sustainable development [12]. The framework of sustainable development permits a socio-political understanding of how people relate to the world to usefully inform humanity’s future [11]. The U.N. Millennium Development Goals convey the current focus of sustainable development, with an emphasis on being able to address complicated global challenges while meeting humanity’s needs and protecting the environment [12].

The findings of the research claim that the authenticity of local food plays a significant role as the key attribute that makes food tourism sustainable and enhances tourists’ satisfaction levels [6]. However, no matter how well food expresses local identity and is a sustainable tourism resource, it will only shine more when there are tourists who seek to visit destinations to experience and consume it. Within food tourism, some studies that combine food and tourism have begun to draw attention and examine food from different perspectives, such as local food consumption and value [13], evaluating tourists’ food experiences [14,15], and challenges concerning food tourism [16].

Previous research has found that individuals tend to avoid eating new foods that are unfamiliar (e.g., ethnic foods) and try to eat foods that they have already experienced [17]. In other words, tourists want to enjoy new cultures and unique tourism experiences in a new environment; however, they may not feel comfortable eating unfamiliar foods due to perceived risks. Research about perceived risk among tourists has been steadily conducted in various contexts, including perceived risk in decision-making [18,19,20,21], border tourism [22], coastal tourism [23], ecotourism [24,25], and the sharing economy [26,27]. Perceived risk can be an obstacle for tourists to participate in tourism activities and can influence the decisions that are made by tourists [28]. Efforts have been made to promote tourism activities by identifying the perceived risks of tourists and making efforts to reduce these risk factors. However, research to understand what risk factors tourists perceive when visiting other countries to experience their unique food is currently insufficient. Therefore, this study’s purpose was to discover the risk factors that tourists perceive when they travel to experience food culture and seek to experience food itself in a new environment. The theory of planned behavior (hereafter TPB) is well-known as a theory that can forecast tourists’ behavioral intentions and has been used in the following diverse settings: food delivery services [29], well-being food [30], and tourism destinations [31]. It has been demonstrated to help predict tourists’ behavioral intentions. Thus, the extended theory of planned behavior (hereafter ETPB) is a highly employed model in the tourism and hospitality fields and was adapted for this study.

Three research questions guided this study. What risk factors are perceived by tourists in food tourism? What actions do tourists take as a strategy to reduce these perceived food tourism risk factors? What are the structural impact relationships between those risk factors, specific variables of TPB, and coping behavior? Although there is sufficient evidence on the links among perceived risk and three TPB variables (i.e., attitude, perceived behavioral control, and subjective norm), relationships among these constructs need further examination from the perspective of food tourism. Moreover, most of the existing research concerning perceived risk mainly focuses on the influence perceived risk has on tourists’ behavior. Although perceived risk has multidimensional characteristics, these studies have limitations, such as measuring the variable as a single dimension. In addition, research on what actions tourists will take to reduce the risk factors perceived when participating in tourism activities is insufficient. Thus, this study’s findings will serve to redefine relationships among the constructs being studied and shed light on building marketing strategies for tourism managers who attempt to develop a destination as a sustainable food tourism destination.

2. Literature Review

2.1. Perceived Risk

Sustainable tourism has been studied in various ways in tourism research since the emergence of initiatives and proposals aimed at realizing a new concept related to tourism called sustainable tourism [32]. The UNWTO [33] defined the concept of sustainable tourism as “Tourism that takes full account of its current and future economic, social, and environmental impacts, addressing the needs of visitors, the industry, the environment, and host communities” (p. 12). The risk factors perceived by tourists when visiting tourism destinations and the concept of sustainability cannot be separated. There will be no economic benefit to returning to the destinations without tourists visiting the places, and tourists will not be able to experience the traditional food and culture of the places that they can access by visiting a specific area. In addition, natural resources managed by tourists’ income will inevitably suffer losses. Therefore, it can be said that there is a close connection between understanding what risk factors tourists perceive when they visit Jeju Island for food tourism and making the area a sustainable tourist destination. Perceived risk has been described as a person’s instinctive judgment of risk and relates to the individual decisions a person makes when evaluating risky experiences or activities that can occur [34]. When making decisions, people anticipate the potential for uncertainty or negative outcomes, generally taking actions to decrease risk and potential questionable outcomes [35,36].

Different dimensions of perceived risk were identified by early scholarship on this topic. Kaplan et al. [37] validated specific perceived risk dimensions that had been previously proposed: physical, psychological, social, financial, and performance risk. Different perceived risk dimensions have been explored in various settings and by numerous tourism researchers. One common type of perceived risk that is inherent to and common in tourism contexts is physical risk, as it has been identified and acknowledged by many researchers [36,38,39,40,41,42]. Performance risk describes the consumption product malfunctioning and not performing as anticipated [43], and in a tourism context, it refers to concern about purchasing a tourism product [44]. This dimension of perceived risk has also been identified by prior tourism researchers [41,45] and has been applied in consumption contexts related to tourism experiences [27,46,47,48]. Communication risk was a dimension of perceived risk added more recently by Han [49] and refers to potential negative consequences arising from language issues or communication challenges related to tourism. It was later tested by An [22] in a study about tourism at border destinations and confirmed to be a relevant perceived risk dimension in the context of border tourism. More recently, Emami and Ranjbarian [50] confirmed communication risk as one perceived risk dimension in their study about perceived risk and tourism in Iran.

2.2. Theory of Planned Behavior and Extended Theory of Planned Behavior

The theory of reasoned action considered how attitude influenced behavior, and the importance of intention was noted as a critical concept that can help to understand what drives a person to implement specific behaviors [51]. Ajzen’s theory of planned behavior (TPB) puts forth that subjective norm, attitude toward a behavior, and perceived behavioral control effect intention, which then influences real behavior [51,52]. TPB expanded on the theory of reasoned action by including a variable for perceived behavioral control, and the components of this theory can account for behavioral intentions, although the relevance of each component will vary depending on the specific circumstances [52]. TPB has been applied in many contexts. Armitage and Conner [53] completed a meta-analysis of 185 TPB studies and found that TPB was predictive of both behavior and intention. More recently, a review of TPB studies in tourism, hospitality, and leisure contexts noted that consumer behavior was the research context wherein the highest number of studies in the field had been conducted [54]. Similarly, the largest number of studies reviewed were conducted in Western countries [54]. Additionally, sustainability and TPB within this field have only become more researched in recent years [54]. Thus, further researching the TPB model with a focus on behavioral intention in an Eastern context would be a useful addition to the literature. Perhaps more importantly, examining TPB in relation to sustainability in a food tourism context is relevant given that improvement in tourism practice remains relevant for sustainability in tourism [55].

Extensions or modifications of the TPB model have come to be commonly applied in tourism studies. Su et al. [56] discovered that attitude and perceived behavioral control affected behavioral intention in a food tourism context. Behavioral intention was not influenced by subjective norms, according to these researchers, but food travel motivation had a significant influence on that variable. An expanded TPB was used by Abbasi et al. [57], and perceived behavioral control and perceived value significantly influenced revisit intention in the study. However, attitude, subjective norm, and perceived risk did not significantly influence revisit intention.

Troise et al. [58] examined online food delivery services and behavioral intention, testing the TPB framework as part of their analysis. The researchers noted that behavioral intention was affected by attitude, perceived behavioral control, and subjective norms. Additionally, it was found that subjective norms influenced understanding of COVID-19-related risks, and this risk perception influenced behavioral intention. Perception of risks did not significantly affect attitude in the study. Seong and Hong [31] and Seong et al. [59] used extended TPB (ETPB) models to better understand COVID-19 risk and national park visitation. Both studies found that COVID-19 risk perceptions influenced attitudes, perceived behavioral control, and subjective norms. Seong and Hong [31] discovered that these three variables affected visit intention. Seong et al. [59] noted that perceived behavioral control and subjective norms influenced coping behavior. Zeydan and Gürbüz [60] also used ETPB to study travel intentions related to COVID-19. In this study, the researchers learned that perceived behavioral control, attitude, and subjective norms affected travel intentions. Additionally, perceived COVID-19 risk influenced behaviors related to risk reduction, which in turn influenced the intention to travel in this study.

2.3. Coping Behavior

Recent research has shed light on the coping behaviors taken to reduce actual or perceived risks in tourism and leisure environments. Examining alternatives and altering behaviors by going to areas that are not crowded or to other areas was related to coping behaviors in a leisure context [61]. Manning [62] used a congestion model to better understand tourists’ behavior, as tourists who knew about site congestion at tourism attractions were willing to avoid the attraction or visit other less crowded areas. These findings connected behavioral changes to what people engage in as coping behaviors. Tourists have also changed the time of visitation or visited less crowded areas [62] to deal with the issue of crowding. It has been noted that tourists will take actions to determine behavioral guidelines for themselves [63]. Tourists have been known to modify their behavior due to their risk perceptions, and this can take the form of evading areas that are crowded as well as making decisions concerning who to travel with [64].

The destination choices of tourists have also been influenced by the sources of information that they used, which was a strategy that could serve to decrease the risk perceptions of tourists [65]. Making behavioral changes and initiating corrective measures [66], as well as avoiding crowded spaces [64], have been identified as ways that tourists and consumers seek to adjust behaviors to reduce risk. In tourism, individuals derive their own corrective behaviors due to risk [67]. Dayour et al. [67] examined risk perceptions regarding the use of smartphones and noted that cognitive and proactive behavioral measures were used by backpackers. Negotiation strategies are another strategy for tourists to lessen the factors that serve as behavioral constraints [68]. Consumer responses to risk as well as what produces behaviors oriented towards risk reduction can be further researched [69].

2.4. The Relationships between Constructs

2.4.1. Perceived Risk and TPB Variables

Perceived risk has been related to TPB variables in several prior studies. Lee [70] studied variables affecting the use of internet banking. Different types of risk (i.e., performance, time, financial, and security risks) influenced attitudes in the study, and social risk was found to affect social norms. Intention was influenced by social norms, perceived behavioral control, and attitude in the study. Sanayei and Bahmani [71] studied risk perceptions and consumer acceptance of internet banking. Various risks (i.e., social, time, performance, financial, and security risks) negatively influenced attitudes. Additionally, security and financial risk negatively influenced the intention to use. Xie et al. [72] studied predictors for e-government adoption. The researchers found that trust in e-government could reduce the perceived risk of reusing e-government services. Additionally, perceived risk influenced attitudes and perceived behavioral control. Ha [73] noted that attitude and subjective norms positively affected online shopping intention. Perceived risk negatively affected online shopping intentions among consumers in that study. Tobias-Mamina and Maziriri [74] found that perceived risks (i.e., performance risk, financial risk, and security risk) influenced the inclination to utilize online banking services. More recently, Lim and An [30] applied the TPB model to examine consumers’ purchase intentions for well-being food. They found that the constructs of attitude, subjective norms, and perceived behavioral control showed a significant contribution to the intention to purchase well-being food among Korean consumers. These studies highlighted the relevance of risk to both consumers and citizens.

2.4.2. TPB Variables and Coping Behaviors

Coping behaviors and approaches to reduce risk have been related to TPB in a few recent studies. Seong and Hong [31] studied COVID-19 risk perception and national park visitation. COVID-19 risk perception negatively affected subjective norms, perceived behavioral control, and attitude in the study. The three variables positively influenced visit intention, which in turn influenced risk reduction behavior along with COVID-19 risk perception. Seong et al. [59] applied the framework of TPB to examine factors that influenced national park visitation behaviors. Risk perception pertaining to COVID-19 positively influenced subjective norms, perceived behavioral control, and attitude in the study. In turn, perceived behavioral control and subjective norms affected coping behavior. Attitude had no significant influence on visit intention, according to this research.

Jomnonkwao et al. [75] used TPB to explore the inclination to pay for accident risk reduction. The researchers found that the driver’s attitude, perceived behavioral control, and subjective norm affected the behavioral intention, which in turn influenced the driver’s willingness to pay. Thus, using TPB in this study helped to explain attitudes towards the payment of accident risk reduction, demonstrating TPB’s utility in a risk reduction context. Zeydan and Gürbüz [60] applied ETPB to the context of tourists’ travel intentions in Turkey. These researchers found that COVID-19 risk perception affected risk reduction behavior. The behavior to reduce risk then affected travel intention. Recent studies have highlighted the reduction of risk perceptions as applied along with the TPB model as a useful way of seeking to understand tourists’ coping behaviors and risk reduction strategies.

2.4.3. Coping Behaviors and Intention to Travel

Coping behaviors have been related to consumer or tourist intentions in a few recent research studies. Hsieh et al. [76] studied sustainable transportation and noted empirical support for the notion that coping planning influences behavioral changes in relation to changes in car usage and increased use of pro-environmental modes.

Many recent studies have investigated coping behaviors in relation to COVID-19 and how this influences travel. Seong et al. [59] found that coping behavior influenced sustainable intention to visit in their study about national park visitors. Morar et al. [77] studied travel behaviors, tourists’ personalities, and the pandemic. These researchers noted that travel fear led to travel avoidance. Emotional coping was associated with safety in the study. Successful coping behaviors, which can incorporate problem solving, resulted in cautious travel intentions among study participants.

A study by Melania [78] found that humor as a coping mechanism served as a mediating variable that negatively influenced behavioral intention to travel. Kurniawati [79] studied strategies to reduce risk and travel intention. It was found that strategies to reduce financial risk and destination trust positively affected visit intention. Zheng et al. [80] also researched COVID-19 travel behaviors. These researchers discovered that different forms of coping enhanced resilience, which resulted in cautious travel. In other words, greater coping ability resulted in the intention to travel with caution.

Han et al. [81] researched tourists’ behavior before and following the COVID-19 pandemic. These researchers found that coping behaviors influenced psychological resilience, which in turn influenced travel intentions before and after the pandemic. Such contemporary research indicates that coping behaviors can influence tourists’ intentions to travel. Based on the reviewed literature above, 13 hypotheses were developed (see Figure 1).

H1. 

Physical risk has a negative effect on attitude.

H2. 

Physical risk has a negative effect on subjective norms.

H3. 

Physical risk has a negative effect on perceived behavioral control.

H4. 

Communication risk has a negative effect on attitude.

H5. 

Communication risk has a negative effect on subjective norms.

H6. 

Communication risk has a negative effect on perceived behavioral control.

H7. 

Food performance risk has a negative effect on attitude.

H8. 

Food performance risk has a negative effect on subjective norm.

H9. 

Food performance risk has a negative effect on perceived behavioral control.

H10. 

Attitude has a positive effect on coping behavior.

H11. 

Subjective norms have a positive effect on coping behavior.

H12. 

Perceived behavioral control has a positive effect on coping behavior.

H13. 

Coping behavior has a positive effect on the intention to travel.

3. Methodology

3.1. Construct Measurement

A survey questionnaire was developed to explore the relationships among perceived risk, three TPB variables, and coping behavior. This survey was shared with Chinese tourists that had considered traveling to Jeju Island in South Korea for food tourism. For the perceived risk variable, 15 items were drawn from prior research [14,82]. The coping behavior variable consisted of three items that were modified from previous studies [83,84]. Finally, the items measuring the three TPB variables were drawn from past research [59,85]. The questions in the survey questionnaire were measured on a 5-point Likert scale. The scale ranged from 1 = strongly disagree to 5 = strongly agree.

3.2. Sampling and Data Collection

First, the questionnaire was developed in English. Then, translation to Mandarin Chinese took place, given that this study’s participants were Chinese. A researcher who is fluent in both Mandarin and English completed the translation. Several professors and graduate students majoring in tourism were contacted to review the translated questionnaire to ensure linguistic validation. Given the feedback received in this review process, some modifications were made to the questionnaire, and it was then pilot tested with 30 Chinese tourists. Any issues related to the grammar or wording of the questionnaire were acknowledged, and the final questionnaire was made based on what was learned.

Jeju Island is one of the most famous tourism destinations in South Korea. International tourists to Jeju decreased during the period of COVID-19. However, it has increased gradually during the recovery period. In 2023, approximately 243,238 Chinese tourists visited Jeju as of September, and Chinese tourists accounted for the largest portion of international tourists [86]. After the COVID-19 pandemic, the city of Jeju has sought to re-attract Chinese tourists, which account for the largest portion of international tourists visiting Jeju Island. To accomplish this, it is necessary to first understand the risks Chinese tourists perceive when they consider visiting Jeju Island. Since traditional food in Jeju, such as Korean BBQ and seafood, is popular with international tourists, the population target for this study was Chinese tourists who considered visiting Jeju and who were willing to try traditional foods on Jeju Island. Thus, the study population was Chinese tourists aged 18 and over living in China who are potential food tourists who considered traveling to Jeju Island for food tourism. The study population was Chinese tourists aged 18 and over who considered traveling to Jeju Island. An online questionnaire was developed, and Sojump, a famous Chinese survey platform, was employed. Then, the online survey link was distributed through WeChat, the most well-known social media application in China, to Chinese tourists. Data were collected beginning on 15 March 2023, for a period of two weeks, generating 320 completed responses. After removing 17 incomplete responses, 303 completed responses remained to be used for further analysis.

3.3. Data Analysis

SPSS (Statistical Package for Social Sciences) 27.0 was used in order to complete the data analysis. The data analysis process specifically included Exploratory Factor Analysis, Confirmatory Factor Analysis, and Structural Equation Modeling utilizing AMOS 27.

4. Results

4.1. Descriptive Statistics

The responses analyzed for this study totaled 303 valid responses. Information from respondents pertaining to demographic characteristics such as age, gender, monthly income, employment, frequency of food tourism experiences in the past five years, and who accompanied them was analyzed with the SPSS 27.0 program (see

Table 1. Demographic information of the respondents.

Variables	Item	N	%
Gender	Male	139	45.9
	Female	164	54.1
Age	20s	75	24.8
	30s	98	32.3
	40s	66	21.8
	50s	36	11.9
	60s	28	9.2
Monthly income (USD)	1000 and less	56	18.5
	1001–2000	117	38.6
	2001–3000	91	30.0
	Over 3000	39	12.9
Employment or professional status	Student	21	6.9
	Office worker	118	38.9
	Professional	46	15.2
	Self-employed	72	23.8
	Other	46	15.2
Number of food tourism experiences	1–2	64	21.1
	3–4	93	30.7
	5–6	87	28.7
	More than 7 times	59	19.5
Destination of food tourism	Domestic	102	33.7
	International	86	28.4
	Both	115	37.9
Accompany	Family members	76	25.1
	Non-Family members	198	65.3
	Alone	29	9.6
	Total	303	100

). More females (54.1%) responded than males (45.9%). The age group most represented was 30 s (32.3%). Less than half of the respondents were office workers (38.9%), and in terms of the income range of respondents, $1000 to $2000 (38.6%) was the category selected by most respondents. Among the respondents, 3–4 food tourism experiences (30.7%) were the most represented number of food tourism experiences in the past five years. About 38% of respondents participated in food tourism activities domestically and abroad. For those respondents who experienced food tourism, friends and colleagues accounted for the largest percentage (65.3%) of those who accompanied participants during their experiences.

4.2. Exploratory Factor Analysis (EFA)

Exploratory factor analysis (EFA) with principal component analysis was performed in order to better understand the variables that could be used for further analysis as well as the perceived risk dimensions that could be further analyzed. The results, as indicated in

Table 2. Exploratory factor analysis results.

Variables and Items	Factor Loading	Eigen Value	Variance Explained	Cronbach’s α
Physical risk		4.356	14.050	0.911
Potential health problems	0.767
Get sick from food	0.766
Possibility of contacting infectious diseases while dining	0.746
Possibility of man-made violent events	0.714
Possibility of public security incidents	0.644
May be sick on the trip (e.g., pneumonia)	0.625
Communication risk
The inconvenience of using other languages	0.873	2.941	9.488	0.861
Difficulty in ordering food	0.841
Difficulty in understanding the menu	0.829
Communication problems while dining	0.719
Food performance risk
Food arrangements are not as expected	0.886	2.675	8.630	0.785
Food taste is unsatisfactory	0.880
Food tourism is unable to meet the requirements of cultural experiences	0.876
Attitude
I like to visit Jeju for food tourism	0.816	2.437	7.862	0.744
happy to visit Jeju for food tourism	0.755
Visiting Jeju for food tourism is worthwhile	0.668
Visiting Jeju for food tourism will have good outcomes	0.592
Subjective norm
My family thinks positively about my trip to Jeju for food tourism	0.733	2.382	7.685	0.893
My friends think positively about my trip to Jeju for food tourism	0.717
My family will want me to visit Jeju for food tourism	0.709
My friends will want me to visit Jeju for food tourism	0.525
Perceived behavioral control
I can visit Jeju for food tourism whenever I want	0.742	2.284	7.366	0.854
I am financially able to afford to visit Jeju for food tourism	0.645
I have enough time to visit Jeju for food tourism	0.640
I can easily find the information about visiting Jeju for food tourism	0.618
Coping behavior
Search for information about food tourism in Jeju	0.798	2.164	6.982	0.743
Choose a famous restaurant	0.747
Learn to speak the language for a simple conversation	0.661
Intention to travel
Will make an effort	0.779	1.994	6.442	0.713
Intend to visit	0.771
Want to visit	0.577

, showed good representativeness (KMO = 0.895), and Bartlett’s sphericity test was significant (χ²(465) = 4752.590, p < 0.001), indicating that the data were appropriate for analysis. Among 15 items measuring perceived risk, two items (e.g., “I worry that I might not get value for money” and “It would not be a good idea to spend my money on buying some food I do not know”) were excluded due to the communalities being lower than 0.3. Principal component analysis for perceived risk demonstrated three dimensions with an eigenvalue of 1 or more. The total variance explained was 68.51%. Three dimensions of perceived risk were extracted and labeled: “Physical risk,” “Communication risk,” and “Food performance risk.” Cronbach alpha coefficients were examined to assess the reliability of the perceived risk dimensions, and the values were 0.91, 0.86, and 0.79, respectively, for the three perceived risk dimensions.

One factor was obtained during factor analysis with an eigenvalue greater than 1 (eigenvalue = 2.44) for the attitude variable, and the Cronbach alpha value was 0.74. Subjective norms and perceived behavioral control variables were also extracted as one single factor. The eigenvalues were shown to be greater than 1 (eigenvalue = 2.38 and 2.28) and the Cronbach alpha amounts were greater than the cut-off of 0.7 (0.89 and 0.85). The principal component analysis for the coping behavior variable was extracted as one single factor. The eigenvalue was more than 1 (eigenvalue = 2.16) and the Cronbach alpha value was 0.74. The "Intention to travel" variable was extracted as a single factor. The eigenvalue was also greater than 1 (eigenvalue = 1.99), and the Cronbach alpha value was 0.71.

4.3. Confirmatory Factor Analysis (CFA)

Confirmatory factor analysis (CFA) was performed to assess the appositeness of the proposed measurement model, as shown in

Table 3. Confirmatory factor analysis results.

Variables and Items	Standardized Loading	S.E.	C.R.	CR	AVE
Physical risk
Potential health problems	0.792
Get sick from food	0.756	0.067	13.974	0.902	0.605
Possibility of contacting infectious diseases while dining	0.634	0.074	11.320
Possibility of man-made violent events	0.802	0.070	15.052
Possibility of public security incidents	0.788	0.073	14.723
May be sick on the trip (e.g., pneumonia)	0.791	0.074	12.538
Communication risk
The inconvenience of using other languages	0.726
Difficulty in ordering food	0.706	0.128	10.250	0.883	0.626
Difficulty in understanding the menu	0.902	0.124	12.061
Communication problems while dining	0.886	0.131	11.979
Food performance risk
Food arrangements are not as expected	0.857
Food taste is unsatisfactory	0.870	0.058	18.054	0.894	0.737
Food tourism is unable to meet the requirements of cultural experiences	0.848	0.057	17.580
Attitude
I like to visit Jeju for food tourism	0.735
happy to visit Jeju for food tourism	0.809	0.157	8.516	0.873	0.632
Visiting Jeju for food tourism is worthwhile	0.814	0.116	7.521
Visiting Jeju for food tourism will give me good outcomes	0.819	0.127	8.190
Subjective norm
My family thinks positively about my trip to Jeju for food tourism	0.801
My friends think positively about my trip to Jeju for food tourism	0.777	0.086	12.071	0.858	0.610
My family will want me to visit Jeju for food tourism	0.780	0.086	12.120
My friends will want me to visit Jeju for food tourism	0.743	0.087	11.616
Perceived behavioral control
I can visit Jeju for food tourism whenever I want	0.835
I am financially able to afford to visit Jeju for food tourism	0.807	0.103	8.723	0.886	0.660
I have enough time to visit Jeju for food tourism	0.862	0.111	10.320
I can easily find the information about visiting Jeju for food tourism	0.744	0.097	10.164
Coping behavior
Search for information about food tourism in Jeju	0.845
Choose a famous restaurant	0.759	0.090	10.278	0.821	0.605
Learn to speak the language for a simple conversation	0.725	0.076	9.809
Intention to travel
Will make an effort	0.879
Intend to visit	0.730	0.120	8.460	0.830	0.621
Want to visit	0.746	0.146	8.518
χ2 = 737.658, df = 406, CMIN/DF = 1.817, GFI = 0.961 AGFI = 0.930, NFI = 0.951, CFI = 0.926, RMR = 0.034, RMSEA = 0.052

Note. p < 0.001.

. Results indicated acceptable model fit χ²(406) = 737.658, p < 0.05, goodness of fit index (GFI) = 0.96, comparative fit index (CFI) = 0.93, adjusted goodness of fit index (AGFI) = 0.93, normed fit index (NFI) = 0.95, root mean square error of approximation (RMSEA) = 0.05, and root mean square residual (RMR) = 0.03. Moreover, this study evaluated average variance extracted (AVE), composite reliability (CR), convergent validity, Cronbach’s Alpha, and discriminant validity for the latent variables. As shown in Table 3, all measurement concepts significantly corresponded to observed variables because the standardized loading estimates (>0.50) demonstrated that construct validity was satisfactory [87]. Results demonstrated that the average variance extracted (AVE) for all constructs ranged from 0.61 to 0.74. This is more than 0.5, signifying acceptable convergent validity. In terms of construct reliability for the measurement constructs, this ranged from 0.82 to 0.90, which was above recommended thresholds for composite reliability (>0.60) [87].

To evaluate discriminant validity, the AVE values for variables were examined by relating the values with the coefficient of determination among the equivalent constructs as residual latent variables. The AVE values were greater than the R squared values involving all paired constructs, thus denoting endorsement of discriminant validity. As shown in

Table 4. Discriminant validity test of the measurement model.

Variable	PR	CR	FP	AT	SN	PBC	CB	BI
PR	0.737
CR	0.145 (0.021) *	0.605
FP	0.102 (0.010) *	0.518 (0.268) ***	0.621
AT	−0.182 (0.033) *	−0.542 (0.294) ***	−0.482 (0.232) ***	0.660
SN	−0.281 (0.079) ***	−0.594 (0.353) ***	−0.623 (0.388) ***	0.693 (0.480) ***	0.601
PBC	−0.152 (0.023) *	−0.493 (0.243) ***	−0.479 (0.229) ***	0.418 (0.175) ***	0.535 (0.286) ***	0.632
CB	0.395 (0.156) ***	0.526 (0.277) ***	0.468 (0.219) ***	0.768 (0.590) ***	0.706 (0.498) ***	0.453 (0.205) ***	0.605
BI	−0.246 (0.021) ***	−0.210 (0.044) **	−0.309 (0.095) ***	0.349 (0.122) ***	0.372 (0.138) ***	0.344 (0.118) ***	0.359 (0.129) ***	0.626

Notes: (1) PR: Perceived Risk; CR: Communication Risk; FP: Food Performance Risk; AT: Attitude; SN: Subjective Norm; PBC: Perceived Behavioral Control; CB: Coping Behavioral; BI: Behavioral Intention. (2) The bold diagonal elements are the square root of the AVE. (3) Below the diagonal line is the correlation value between the constructions, and () is the correlation coefficient between the constructions. (4) * p < 0.05, ** p < 0.01, *** p < 0.001.

, all AVE for constructs exceeded the cut-off level of 0.5 [88]. Thus, the eight factors of the measurement model were verified, given the confirmation of strong measurement properties.

4.4. Testing Hypotheses

Table 5. Standardized regression weights and model testing.

Hypothesis	Path	Standardized Path Coefficients	t-Value	Results
H1	PHY → ATT	−0.497	−5.497 ***	Supported
H2	PHY → SN	−0.261	−4.760 ***	Supported
H3	PHY → PBC	−0.362	−4.177 ***	Supported
H4	COM → ATT	−0.360	−3.973 ***	Supported
H5	COM → SN	−0.115	−1.847	Rejected
H6	COM → PBC	−0.356	−3.670 ***	Supported
H7	FP → ATT	−0.329	−4.286 ***	Supported
H8	FP → SN	−0.186	−3.112 ***	Supported
H9	FP → PBC	0.064	−1.105	Rejected
H10	ATT → CB	0.536	6.898 ***	Supported
H11	SN → CB	0.356	5.222 ***	Supported
H12	PBC → CB	0.309	3.384 ***	Supported
H13	CB → BI	0.337	5.523 ***	Supported
χ2 = 806.301, df = 418, CMIN/DF = 1.929, GFI = 0.948, AGFI = 0.919, NFI = 0.937, CFI = 0.918, RMR = 0.040, RMSEA = 0.056

Note. (1) PR: Perceived Risk; CR: Communication Risk; FP: Food Performance Risk; AT: Attitude; SN: Subjective Norm; PBC: Perceived Behavioral Control; CB: Coping Behavioral; BI: Behavioral Intention. (2) *** p < 0.001.

shows the results of the 13 hypotheses tests using Structural Equation Modeling (SEM) regarding consumers’ travel intentions to Jeju Island for food tourism. The current research model showed a strong model fit: χ²(418) = 806.30, p < 0.01, GFI = 0.948, AGFI = 0.919, NFI = 0.937, CFI = 0.918, RMR = 0.040, RMSEA = 0.056. All indices were either at satisfactory levels or close to the cut-off level. Hypothesis 1 proposed that a significant negative relationship between physical risk and attitude would exist, and this was supported (β = −0.497, p < 0.001). The results demonstrated that physical risk had a negative effect both on subjective norm and perceived behavioral control (β = −0.261, β = −0.362; p < 0.001). Thus, Hypothesis 2 and Hypothesis 3 were supported. Hypothesis 4 proposed that communication risk would negatively influence attitude, and this was also confirmed as statistically significant (β = −0.360, p < 0.001). Hypothesis 5 indicates that communication risk will have a negative effect on subjective norms. However, results demonstrated the relationship between communication risk and subjective norm was not statistically significant (β = −0.115, p > 0.05). Hypothesis 6 was confirmed as there was a significantly negative relationship found between communication risk and perceived behavioral control (β = −0.356, p < 0.001). Hypotheses 7 and 8 propose that food performance risk would negatively influence attitudes and subjective norms. Results indicated a significant and negative relationship between both variables (β = −0.329, β = −0.186; p < 0.001). Therefore, both Hypotheses 7 and 8 were supported. However, a significant statistical relationship was not found between food performance risk and perceived behavioral control. Thus, Hypothesis 9 was not supported (β = −0.064, p > 0.05). Three TPB variables and coping behavior showed a positive relationship (β = 0.536, β = 0.356, β = 0.309; p < 0.001). Therefore, Hypotheses 10–12 were all supported. Finally, Hypothesis 13 proposed that coping behavior would positively influence travel intention. Results demonstrated that the relationship between coping behavior and travel intention was significant (β = −0.337, p < 0.001), and Hypothesis 13 was supported.

This study specifically examined the interest of Chinese tourists concerning visitation to Jeju Island for food tourism at the time of the end of the COVID-19 pandemic. This research identified three different perceived risk dimensions: physical risk, communication risk, and food performance risk. This answered the first proposed research question because these were the dimensions of perceived risk identified in a food tourism setting. Physical risk is related to accidents or physical threats that tourists may experience when visiting Jeju Island for food tourism; communication risk is a risk factor related to communication issues that tourists may experience when ordering food; and food performance risk is a food-related risk that tourists may face when food consumption fails to perform as expected. To examine the structural relationships between risk dimensions, three variables of TPB, coping behavior, and intention to travel, the 13 hypotheses were established and tested, and 11 were accepted.

5. Conclusions

Risk factors perceived by tourists act as decisive obstacles to choosing tourist destinations and can serve as negative factors pertaining to tourism as an industry [36]. However, tourism studies research identifying food-related perceived risk can be considered insufficient compared to the growth rate of food tourism. Therefore, this study sought to research sustainable food tourism by identifying risk factors perceived by Chinese tourists and what actions they take as a strategy to reduce risk factors. The results of the socio-demographic information of respondents revealed that there were a little more female respondents in terms of gender, but there was no significant difference in response rates between men and women. Respondents who are in their 20 s or 30 s accounted for the largest age groups in this study, seeming to indicate that younger Chinese tourists are more willing to travel to Jeju than older tourists. More than half of the respondents were found to be willing to travel with non-family members (i.e., friends and colleagues) when considering a visit to Jeju Island. Respondents indicated that they have experienced food tourism between three and six times. The number of respondents who experienced international food tourism was relatively high, which can be interpreted as having high interest in and experience in international food tourism.

Results from this study demonstrated a strong and significant relationship between perceived risk and the TPB variables attitude, perceived behavioral control, and subjective norm, a finding in line with prior research [59]. To be more specific, the significant negative coefficient path between physical risk and attitude, perceived behavioral control, and subjective norm was confirmed. This indicated that the higher the perception of physical risk was when tourists intended to participate in food tourism, the more negative their attitude toward food tourism was, and the more negative their perceived behavioral control over food tourism also became. This result is supported by findings from past research, which examined perceived risk and TPB variables [59,89,90]. Rick perception’s negative influence on attitude, perceived behavioral control, and subjective norm variables was noted in this study. Chien et al. [59] also confirmed a negative association concerning perceived risk and perceived behavioral control. However, the relationships among communication risk and the three TPB variables showed different results. First, it was confirmed that communication risk had a negative relationship between perceived behavioral control and attitude, but it was noted that it did not significantly influence subjective norms. In other words, when tourists intend to participate in food tourism activities, a higher communication risk can lead to negative attitudes and lower confidence toward food tourism. Interestingly, communication risk showed no effect on subjective norms. Subjective norms can be seen as not being directly impacted by communication risk. Perhaps the potential for communication challenges or the opinions of others regarding food tourism activities may be viewed as not being directly related to one’s ability. Food performance risk showed a negative relationship with attitude and subjective norms. This indicated that the more tourists feel that the food may be different from what they think, the more they feel that they do not want to engage in food tourism activities or that their acquaintances will have negative opinions about their food tourism experience. However, food performance risk showed no significant relationship with perceived behavioral control. Loh and Hassan [91] found support for food performance risk resulting in a negative attitude, as their study confirmed that tourists who perceived a higher risk of food-related risk had more negative attitudes toward purchasing products from food trucks. Choi et al. [90] also found that perceived risk (in this case, hygienic, environmental, and health risk) showed a negative relationship with attitude in the setting of street food.

The identification in this research of three unique perceived risk dimensions (i.e., physical risk, communication risk, and food performance risk) answered the first research question that guided this study because these were the dimensions of perceived risk identified as being relevant in a food tourism setting. This study also found that positive and significant relationships were confirmed between the three variables of TPB and coping behavior. The items used to measure coping behavior served to answer the second research question, as tourists used the following coping behaviors in this study: searching for more information, choosing a famous restaurant, and learning to speak the language for simple conversations. Along with the relationship with the TPB variables, these coping behaviors are important in a food context to better understand how tourists seek to mitigate perceived risks. From this research, it was found that the more positive attitude tourists have toward food tourism activities, the more active tourists are in trying to lower perceived risk. This is in line with prior research findings [59,89,92]. Chien et al. [89] noted that greater perceived control is related to enhanced risk protection behaviors. Among those three variables, perceived behavioral control showed the highest effect on coping behavior, which parallels the findings of previous researchers [59,93]. Lastly, a strong relationship between coping behavior and behavioral intention was identified. It can be interpreted that when strategies are devised to reduce perceived risk, more tourists are willing to engage in food tourism activities. By testing the ETPB model, the three perceived risk dimensions were empirically demonstrated to have an influence on the three TPB variables they were hypothesized to be influencing. The three TPB variables influenced coping behavior, which influenced the intention to travel. These relationships, as described in this section, served to provide the answer to the final research question.

6. Implications, Limitations, and Future Research Directions

6.1. Theoretical Implications

Theoretical implications for this study are as follows: Firstly, the current study validated an ETPB model that served to further understand Chinese tourists’ attitudes and travel intentions to visit Jeju Island for food tourism. This current study confirmed the ETPB framework that was tested and successfully predicted tourists’ behavioral intentions concerning food tourism.

Secondly, the results of previous studies were validated in the current research, given that perceived risk exists in tourist destinations and can be a decisive obstacle to choosing tourist destinations [28]. This was confirmed to be no exception to food tourism. In food tourism, perceived risk has not been given much attention. Thus, this study contributed to filling a gap in the current literature, especially considering the examination of the three dimensions of perceived risk. Thirdly, most studies that have been examined in the field of tourism so far have mainly focused on the influence of perceived risk on certain behavioral intentions. This study has theoretical significance in examining what actions food tourists take as a strategy to reduce perceived risks. Lastly, this study has theoretical implications for finding a way for food tourism to become a more sustainable tourism activity, given that specific perceived risk factors of tourists were examined from the perspective of tourist behavior. Thus, strategies can be devised to reduce the risk factors. Previous studies lacked research on food tourism from a sustainable tourism perspective, but starting with this study, future studies should research ways to manage food tourism more sustainably.

6.2. Practical Implications

This study also offers several practical implications, and these can be beneficial for tourism practitioners as well as local governments. When partaking in food tourism, tourists were discovered to have the highest perception of physical risk. Thus, more efforts should be made to reduce the safety concerns or the risk of adverse results from tourists eating at restaurants. These efforts can be made first by providing accurate information. In this study, respondents indicated that they seek information related to travel areas, food, and safety as an action that will reduce the risk they perceive. Therefore, tourists should be provided with sufficient information on food tourism activities when establishing a tourism plan. Regarding the provision of information related to food and tourism activities, the information that is provided needs to be up-to-date and accurate. For example, when planning to visit a restaurant that was searched on the Internet during travel, it is possible that the restaurant moved, closed, or that the menu and price information of the restaurant have changed compared with the information provided.

In this study, it was found that attitude was the most influenced by risk among the three variables of TPB. In general, the purpose of participating in tourism activities is to heal, feel pleasure, or experience cultural experiences through tourism, so it is necessary to induce a positive attitude among potential tourists. Behavioral control was the least affected by the risks examined in this study, which means that individuals perceive low risk when they feel that the risk factors perceived in relation to food tourism activities are within their control. Therefore, it is necessary to create a food tourism environment that individuals have some control over. Another useful strategy is to produce and provide videos that make it easier for viewers to understand information related to safety, quarantine rules, and countermeasures in case of adverse incidents. When producing a video, not only information can be delivered, but also safety-related guide videos can be produced. Adding information, such as traditional backgrounds or stories, can also be used to add to the information shared and can help promote food tourism.

6.3. Limitations and Future Research Directions

This study has contributed to the literature, but this research also has several limitations. Firstly, data collection was completed using convenience sampling. As a result, the potential number of food tourists involved in the study was limited. Research examining this topic in the future can use more exacting or systematic sampling methods to include participants who consider traveling to destinations for food tourism purposes. Secondly, while this study collected enough responses among participants who considered traveling to Jeju Island for food tourism, the issue of generalizability of the results remains for future consideration. This study sampled only Chinese food tourists. Therefore, the results of the current research convey the risks perceived by Chinese tourists when engaged in food tourism activities. Thus, it will be meaningful to obtain sufficient responses in consideration of the cultural background of respondents in future studies and examine the differences in the risk factors perceived by people from different countries. This study sought to identify risks perceived by tourists when they participate in food tourism, and unlike previous studies, perceived risks were subdivided into three dimensions. In future studies, it is necessary to determine which risk reduction strategies are more weighted by dividing the risk reduction strategies into sub-dimensions. In addition, it will be meaningful to measure the mediating influence perceived risk reduction behavior has concerning the relationship between perceived risk and behavioral intention. Examining this issue in future research can enhance our understanding of the role risk reduction behavior has on perceived risk and behavioral intention.