Formation Mechanisms of Rural Summer Health Destination Loyalty: Exploration and Comparison of Low- and High-Aged Elderly Leisure Vacation Tourists

Destination loyalty is a key indicator of the competitiveness of tourist destinations. Rural summer health leisure vacations for urban elderly (RSHLVUE) tourists span a wide range of ages. Destination operators need to understand the loyalty formation mechanisms of different aged tourists. RSHLVUE tourists were divided into a low-aged group (LA) and a high-aged group (HA) to examine and modify the hypothesis of the relationship between perceived value, tourist well-being, place attachment, and destination loyalty based on affect, behavior, and cognition (ABC) theory. The test results of the measurement model indicate that the HA showed stronger responses in terms of cost value, sense of meaning, and place dependency. The formation mechanism of destination loyalty for the LA is tourist well-being → perceived value → place attachment → destination loyalty, and for the HA is perceived value → tourist well-being → place attachment → destination loyalty. The findings deepen the understanding of destination loyalty among elderly leisure vacation tourists and can guide RSHLVUE destination managers to enhance destination competitiveness.


Introduction
The increasingly aging population and the implementation of the rural revitalization strategy have produced a new leisure vacation industry in China, namely, rural summer health leisure vacations for the urban elderly (RSHLVUE). By the end of 2021, China had 267.36 million people aged 60 and above, accounting for 18.9% of the total population [1]. A comfortable summer temperature for the elderly is around 26-28 degrees Celsius [2] However, most cities on the southeast side of the Heihe-Tengchong Line, which account for more than 90% of China's total population [3], have a high number of uncomfortable summer days [4]. The retired elderly in these cities have a good economic base and considerable free time [5], so they have the demand for summer vacation and leisure. High-altitude mountainous areas are naturally endowed with comfortable temperatures, high vegetation cover, and fresh natural air for developing RSHLVUE, but these mountainous areas are often inaccessible and have no pillar industries to promote their economic development. With the implementation of the rural revitalization strategy, the transportation infrastructure in these mountainous areas has been improved and the conditions for developing RSHLVUE are available.
RSHLVUE has the characteristics of rural, summer, health, and elderly tourism while developing its own unique characteristics as follows: (1) Tourists are mainly elderly people from nearby cities, (2) the stay length of tourists is usually 1-3 months, (3) the purpose of the vacation is for summer health, (4) the operators of destinations are mainly local villagers, and (5) the destination includes a combination of food and accommodation with a low consumption price. RSHLVUE has been the leading industry in some villages in China. costs incurred. Tourist well-being is the sum of positive emotions and feelings of value that tourism activities bring to a tourist [32]. Dimensions of tourist well-being include satisfaction [33], positive emotions [34], sense of meaning [35], and sense of achievement [36]. When elderly tourists feel better about their health, their positive emotions are stronger. Thus, the health experience is an expression of the well-being of RSHLVUE elderly tourists (see Appendix A for details).
When tourists' perceived value of a destination is high, their level of well-being is high as well [37]. Tourists' perceived value has been confirmed to affect their well-being. For example, Zhang et al. concluded that perceived value is an important antecedent of customers' subjective well-being [38]. Liu et al. found that the emotional, functional, and social values of perceived value directly influence well-being [39]. Junaid et al. argued that the quality, emotional, and knowledge values of tourists' perceived value influence well-being through brand love [40]. Li et al. confirmed that tourists' perceived value positively influenced their subjective well-being [41]. Therefore, the first hypothesis of this study is: H1. Perceived value has a significant positive effect on tourist well-being.

Perceived Value and Place Attachment
The concept of place attachment emphasizes the emotional connection between people and places [42]. Scholars consider that place attachment encompasses emotional and social components [43]. In the RSHLVUE context, place attachment refers to the special emotional relationship that elderly leisure vacation tourists have with destination villages based on the physical environment and social relationships [44].
Positive and strong attachments are created by staying in a place for a long period [45]. Lee et al. argued that tourists' perceptions of positive experiences have a positive effect on tourists' post-tour evaluations and help foster place attachment [46]. Wang et al. found that perceived value has a significant positive effect on place attachment [47]. Zhang et al. showed that perceived value is an important antecedent of place attachment and activity attachment [48]. On this basis, the following hypothesis is proposed: H2. Perceived value has a significant positive effect on place attachment.

Perceived Value and Destination Loyalty
Loyalty is a firm commitment to repurchasing a product or service in the future [48]. Destination loyalty is usually assessed from attitudinal and behavioral perspectives [49]. Loyalty is measured in terms of a tourist's intention to revisit or recommend the destination to others [50]. Tourist perception is an important antecedent of destination loyalty [51,52] and has a significant positive effect on tourists' intention to revisit [53]. Liu et al. found that emotional, functional, and social values in perceived value indirectly influence tourist loyalty through the mediating effect of well-being [39]. Lee et al. argued that tourists' perceived values of service and social well-being contribute to positive post-trip loyalty [46]. Thus, we propose that tourists' perceived value is related to destination loyalty as follows: H3. Perceived value has a significant positive effect on destination loyalty.

Tourist Well-Being and Place Attachment
RSHLVUE tourists usually stay in the countryside destination for a month or more. The countryside's good natural environment, comfortable climate, and fresh air enhance the physical health of elderly leisure vacation tourists. The group living and the richness of activities could also eliminate their loneliness and enhance their well-being [54]. They form an emotional attachment to the destination, that is, place attachment [55]. Bogdan et al. revealed a significant positive relationship between place attachment and tourist wellbeing [56]. Vada et al. indicated that tourist well-being acts as a moderating variable for memorable tourism experiences and place attachment [57]. Basu et al. suggested that the relationship between place attachment and well-being is direct and significant [58].  [59]. To this end, the fourth proposition of this research is developed as follows: H4. Tourist well-being has a significant positive effect on place attachment.

Tourist Well-Being and Destination Loyalty
Tourist well-being is the most powerful determinant of positive behavioral intentions. For example, Jamaludin et al. found that relatively stable and happy groups were important in ensuring destination loyalty intentions [60]. Vada et al. concluded that hedonic well-being has a significant impact on revisit intentions and positive word-of-mouth [61]. Kim and Kim argued that psychological well-being significantly influences loyalty [62]. Reitsamer and Brunner-Sperdin revealed that tourist well-being has a significant positive effect on their intention to revisit and their willingness to engage in positive word-of-mouth about a destination [37]. Liu et al. asserted that tourist well-being acts as a mediator between perceived value and destination loyalty, directly influencing destination loyalty [39]. The fifth hypothesis of this study is as follows: H5. Tourist well-being has a significant positive effect on destination loyalty.

Place Attachment and Destination Loyalty
Place attachment is seen as an important push for tourists to destinations and is an important factor influencing tourists' behavioral intentions [63]. The higher the level of place attachment to a tourist destination, the more loyal to it they tend to be [64]. Horakova et al. suggested that consumers who have a strong attachment to a place spread positive word-of-mouth about it regardless of the environment [65]. Gautam concluded that vacation satisfaction, place attachment, and emotional experience are significant predictors of destination loyalty [66]. Han et al. indicated that place attachment indirectly influences revisit intention through attitude, subjective norms, and perceived behavioral control [67]. On the basis of the previous literature, we posit that place attachment contributes to destination loyalty as follows: H6. Place attachment has a significant positive effect on destination loyalty.

Preliminary General Conceptual Framework
The CAB framework in ABC theory considers the consumer decision-making process for a product to be similar to a problem-solving process, with the following steps [68,69].
(1) Consumers develop their own perceptions of goods or services through their own accumulated experience. (2) They form their own feelings (emotions) through the evaluation of these perceptions. (3) They behave accordingly to their emotions. The cognitive dimension in this study is perceived value, the affective dimension is tourist well-being and place attachment, and the behavioral dimension is destination loyalty, and the initial conceptual model M 0 of RSHLVUE destination loyalty is proposed, as shown in Figure 1.

Research Sites
The research site for this study was Zhongyuan Township, Jiangxi Province, China. It is about 100 km from Nanchang, the capital of Jiangxi Province. Its average altitude is 655 m, air humidity is 82.5%, forest coverage is more than 86%, and air has a negative

Research Sites
The research site for this study was Zhongyuan Township, Jiangxi Province, China. It is about 100 km from Nanchang, the capital of Jiangxi Province. Its average altitude is 655 m, air humidity is 82.5%, forest coverage is more than 86%, and air has a negative oxygen ion content of 100,000 ions per cubic centimeter. Its average summer temperature is 20-22 • C, which is 6-10 • C lower than that of Nanchang City. Zhongyuan Township has been developing RSHLVUE for more than 20 years. By 2020, the destination had more than 640 nongjiales (Farmhouse Joy restaurants), with more than 20,000 beds and 2000 employees. The comprehensive income from summer vacation tourism amounted to 130 million yuan. The main tourists come from Nanchang. Thus, Zhongyuan Township provides a convenient venue for researchers to observe, extract, and analyze the mechanisms of destination loyalty formation among elderly tourists.

Survey Procedure
First, to identify the measurement dimensions of perceived value, tourist well-being, place attachment, and destination loyalty, a systematic literature review and the first semistructured interviews were conducted. The first semi-structured interview was conducted from 1-4 July 2021 in Zhongyuan Township with 15 elderly tourists. Each participant was interviewed for about 45 min.
Second, a questionnaire survey was conducted. The survey was carried out by 14 students trained in research methods. Owing to eyesight problems, most elderly people were interviewed verbally for the survey. The survey was conducted in two phases: (1) A pre-survey was conducted from 22-23 July 2021; (2) a formal survey was conducted from 25 July-5 August 2021.
To reduce the effect of common method biases, a questionnaire was designed with three sets of questions in different orders. The Harman one-way test and the error variable control method test were conducted referring to Podsakoff et al. [70] and Meade et al. [71]. In the Harman one-way test, a common method bias exists if a single factor accounting for more than 50% of the variance is present [70]. In the error variable control method, a common method bias exists if the model fit metrics do not become better with the inclusion of a common variance bias latent variable [71].
Finally, the second semi-structured interviews were conducted from 25-31 August 2021. Twenty tourists were interviewed, ten of whom were in the 45-74 age group and ten in the >75 age group. The average interview time for each interviewee was 45 min.

Measures
In this study, the measures of the four main constructs-perceived value, tourist wellbeing, place attachment, and destination loyalty-were drawn from previous research. The items for the five dimensions of perceived value were adapted from the existing literature [72][73][74]. The items of two dimensions related to place attachment were extracted from the existing literature [75,76]. The two items for destination loyalty were drawn from previous studies [77]. The items for emotional experience and sense of meaning in tourist well-being were adapted from previous studies [78,79], and the items for health experience in tourist well-being were derived from research summaries (See Appendix A).
First, to refine the 52 preliminary items measuring the four constructs from literature and summaries, the expert panel method was applied. The expert panel was composed of three professors of health and wellness tourism research. In this study, the Delphi expert consultation was conducted using back-to-back expert anonymous questionnaires, and a combination of online and offline methods was used to collect data. The investigators explained to each invited expert to ensure that they understood the content of the items and the scoring principles. The experts were asked to rate the importance of each initial dimension on a five-level Likert scale. The initial dimension list with the mean of importance assignment > 3.5, the coefficient of variation < 0.25, and the full score ratio > 20% were Behav. Sci. 2022, 12, 367 6 of 22 selected to form the second version of the initial dimension group. Experts were asked to classify the initial items into the second version of the initial dimension list, and they found that 80% of the items were categorized consistently. Thus, the final initial dimension was determined. In the end, nine items were removed.
Second, based on the modification suggestions collected during the pre-survey phase, 16 items were deleted, and 8 items were modified for clarity and comprehensibility. All items were scored using a five-point Likert scale ranging from strongly disagree (1) to strongly agree (5) (See Appendix B). The final measurement items are shown in Table 1.

Respondents' Profile and Common Methodological Bias Test Results
A one-to-one pilot survey was conducted with 100 respondents, and 92 valid questionnaires were returned. A total of 700 questionnaires were distributed during the formal survey phase, and 641 valid questionnaires were returned, with a valid return rate of 91.57%. The demographic characteristics of the respondents are shown in Table 2.
A common methodological bias test was conducted on the formal survey data. The Harman one-way test showed 11 factors, with the major factor accounting for 29.1%. The results of the error variable control method test showed that the model fit metrics did not become better with the inclusion of the common variance bias latent variable. The common methodological bias was confirmed to not have a significant effect on the results of this study.

Differences in the Four Constructs of Different Aged Groups
A one-way ANOVA test was conducted to understand the differences in the dimensions of the four constructs among elderly leisure vacation tourists of different aged groups; p < 0.05 was considered statistically significant. Table 3 shows that the dimensions of the four constructs differed significantly across the aged groups of the leisure vacation tourists. The results of the post hoc LSD test indicate that high-aged elderly leisure vacation tourists show stronger responses in terms of cost value, sense of meaning, and place dependency. The stronger responses of high-aged elderly leisure vacation tourists in the value of cost can be explained in two ways. On the one hand, the older leisure vacation tourists are, the higher the probability of becoming sick. High-aged leisure vacation tourists consider more expenses for medical care [80]. As a result, they prefer the cheaper and more costeffective RSHLVUE. On the other hand, prices rise faster than pensions increase, potentially bringing down their consumption levels [81]. Concerning the sense of meaning, the highaged elderly are often more respected and humbled by low-aged elderly during their travels due to their rich experiences and can therefore develop a stronger sense of meaning [82]. In terms of place dependence, leisure vacation tourists over 80 years old showed the strongest result. This may be due the high-aged elderly leisure vacation tourists preferring familiar environments and being less adaptable to new ones [83].
Based on the above significant differences between low-and high-elderly people, we divided the elderly tourists into a low-aged group (LA) for 45-74 years and a high-aged group (HA) for 75 years and above according to WHO's age classification criteria.
The measurement and structural models of the four constructs of the RSHLVUE destination loyalty formation mechanism were analyzed separately for the low-and highaged groups.

Measurement Model Validation and Correction
To validate the measurement models for the four constructs in this study, construct validity, convergent validity, and discriminant validity were assessed using samples from the entire aged group (EA) (n = 641), the LA (n = 329), and the HA (n = 312). As shown in Table 4, the Cronbach's α for all three sample groups was greater than 0.7 for each construct and its dimensions, and the composite reliability (CR) values were greater than 0.8. This result indicates that the measurement model has good reliability and internal consistency. Kaiser-Meyer-Olkin indices were all greater than 0.7 and Bartlett's spherical test was significant at the 0.001 level, indicating good overall construct validity. The factor loadings for each dimension's items were all greater than 0.5, with the majority exceeding 0.7, and the average variance extracted (AVE) values were all greater than 0.5, indicating high convergent validity. The correlation coefficient between any two dimensions was less than the square root of each dimension's own AVE (see Appendix C), so the measurement model had good discriminant validity. Thus, the reliability and validity of the constructs and their dimensions in the measurement model were satisfied. The M 0 was validated using the data from the LA sample, and paths H1, H3, H4, and H5 were not significant. This result indicates that the effect of tourist well-being on destination loyalty is not significant. However, a strong correlation was found between tourist well-being and destination loyalty [39]. Additional semi-structured interviews were conducted to investigate the reasons for the result from 26-31 August 2021.
From the interviews, we know that the LA are more inclined to be involved in activities and travel around than the HA. The better their physical condition is, the stronger their ability to perceive the various elements of the places they visit. Because of the improved health experience in the tourist well-being construct, LA had a better perceived value.
The LA are in the stage of social role transition, and their sense of emptiness is more intense than that of the HA [84]. A significant proportion of the LA relieve the emptiness by traveling. This means that LA have more anticipation and generally higher positive emotions in the pre-tourism period compared with the HA. This positive emotion facilitates the participation of the LA in various activities during the leisure vacation, in turn affecting the perceived value of the destination. Sreejesh and Ponnam suggested that emotions can influence perceived value [85]. Kiviniemi et al. argued that emotions and perceived value are in an interactive relationship, not a one-way influence [86]. Therefore, this study proposes the following additional hypothesis.

H1 .
Tourist well-being has a significant positive effect on perceived value.
On the basis of M 0 , hypothesis H1 was replaced with H1 to form M 1 . M 1 was examined by the LA sample data, and the results showed that H1 was established. To improve the fit index of the model, M 1 was modified to obtain M 2 as follows. (1) The cost value dimension was removed, as the perceived value to cost value significance level (p-value) was greater than 0.001 and had a factor loading of 0.18 (<0.3). (2) The sense of meaning dimension was removed because the p-value of the tourist well-being to a sense of meaning was greater than 0.001 and a factor loading of 0.29 (<0.3). (3) The association of EME2 with FUV3 and PLD1 with SEV2 was established because these two groups had the highest residual correction indices. The fit indices of models M 1 and M 2 were compared (Table 5), and M 2 could fit the sample data better. Figure 2 shows the influence relationship between the constructs of M 2 and their influence path.    The results showed that the p-values of H1 , H2, and H6 were significant, and the three hypotheses were accepted. Additionally, the p-values of H3, H4, and H5 were not significant, so the three hypotheses were rejected (Table 6). In terms of total variance explained values, 97% of perceived value can be explained by tourist well-being (R 2 = 0.97), 86% of place attachment can be explained by perceived value (R 2 = 0.86), and place attachment explains 32% of destination loyalty (R 2 = 0.32). The mechanism of LA leisure vacation destinations loyalty is that perceived value acts as a mediator of tourist well-being and place attachment, which then indirectly influences destination loyalty through place attachment.

High-Aged Group
The M 0 was validated with data from the HA sample, and the model fit results were good. To further optimize the model, the association of EME2 with FUV3 and PLD1 with SEV2 was established to obtain M 3 . The fitted index of M 3 is better than that of M 0 (Table 7). Figure 3 reveals the influence relationship between the constructs and the influence path. This study also tests H1 , whose model fit index results are not satisfactory. Therefore, the H1 is more suitable for HA. The empirical results showed that the p-values of H1, H4, and H6 were significant, and the three hypotheses were accepted. The p-values for H2, H3, and H5 were not significant, and the three hypotheses were rejected (Table 8). In terms of the total variance explaining the values, perceived value explained 78% of tourist well-being (R 2 = 0.78), tourist well-being explained 86% of place attachment (R 2 = 0.86), and 50% of destination loyalty was explained through place attachment (R 2 = 0.50). Perceived value through the two continuous mediating variables, tourist well-being and place attachment, indirectly affect destination loyalty.

Comparison of the Low-and High-Aged Groups
The results from Sections 4.3 and 4.4show a significant difference between the LA and HA. First, the measurement model differences were demonstrated on the basis of the factor loadings of the two sample groups under the same construct as follows. (1) Perceived value construct: The environmental value dimension and the social value dimension were more important for the LA. The effect of the cost value dimension was only

Comparison of the Low-and High-Aged Groups
The results from Second, in terms of structural models, significant differences were found as follows.

Conclusions
Based on ABC theory, this study is the first to explore the formation mechanism of RSHLVUE destination loyalty by aged grouping of elderly leisure vacation tourists. The results confirm that perceived value, tourist well-being, and place attachment are important antecedents of destination loyalty. However, the role of these antecedents in the formation mechanism of destination loyalty varies across aged groups. The specific findings are as follows.
(1) In the RSHLVUE context, significant differences exist in the dimensions of each construct in the destination loyalty formation mechanism of LA and HA, with the most significant differences in the cost value dimension of the perceived value construct and the sense of meaning dimension of the tourist well-being construct. (2) Data from the LA and HA samples confirmed that place attachment directly affects destination loyalty.
(3) Data from the HA sample confirmed the effect of tourist well-being and place attachment as continuous mediating variables that transmit perceived value on destination loyalty. (4) Data from a sample of the LA confirmed that tourist well-being indirectly influences place attachment through perceived value.

Theoretical Implications
This study makes the following theoretical contributions. First, based on ABC theory, tourist well-being is introduced into the study of the formation mechanism of RSHLVUE destination loyalty, enriching the study of destination loyalty formation mechanism. Existing studies usually consider satisfaction an antecedent of destination loyalty, neglecting the affective factor of tourist well-being. The introduction of tourist well-being in this study allows for a better exploration of the factors influencing the psychological dimension of destination loyalty. In addition to combining the characteristics of RSHLVUE tourists, this study constructs three dimensions of well-being measurement: emotional experience, sense of meaning, and health experience, to provide a scale basis for subsequent research related to the well-being of RSHLVUE tourists.
Second, this study further subdivides elderly leisure vacation tourists by age to explore the loyalty formation mechanism, deepening the study of elderly leisure vacation destination loyalty. Existing studies explored the mechanisms of destination loyalty formation mainly from the perspective of tourists in general, without an in-depth understanding of tourist heterogeneity. This study confirms significant differences in the mechanisms of destination loyalty formation between LA and HA. The findings deepen the understanding of destination loyalty among RSHLVUE tourists and enrich the knowledge related to gerontology.

Management Implications
The local government of the destination can formulate relevant policies to guide the operators to provide services for LA or HA, which will be more beneficial to improve the satisfaction and loyalty of tourists. More importantly, the results and findings of the study provide the following management insights and guidance to destination operators. First, leisure vacation tourists' place attachment to the destination should be fostered. The results showed that destination loyalty of LA and HA was directly influenced by place attachment. Their functional needs for leisure vacation destinations can be met by improving medical support facilities and providing humanized care services to enhance their local dependence. Their sense of identification with leisure vacation destinations can be enhanced by conducting more recreational activities.
Second, more targeted initiatives can be implemented to enhance the RSHLVUE tourist perceived value. For the LA, more focus should be placed on the environmental and social interaction aspects, such as by designing themed accommodations and adding small ornamental plants to enhance their perception of environmental value. By organizing group activities, a platform for interactive friendships among LA is provided to enhance their perception of social value. For HA, the focus on service quality, functional facilities, and price should be increased, such as by providing more cost-effective packages or group purchase discounts to improve the perception of cost value. The provision of facilities such as free medical testing equipment, fitness facilities, and walking aids enhances the perception of functional value. Regular free medical checkups should be held and services such as free errands for medicine should be provided to enhance the perception of service value.
Third, the following differentiated strategies can be used to enhance the well-being of LA and HA. For LA, enhancing their well-being at the initial stage of arrival at the destination is needed. Destination operators focus on enhancing the emotional experience and health experience of the LA. Operators can choose to quickly mobilize the positive emotions and health experience of LA when they first arrive at the destination by organizing small-scale activities or giving them health care gifts. For HA, enhancing the sense of well-being in the middle and later stages of arriving at the destination is necessary, and more attention should be paid to the formation of their sense of meaning. Operators should pay more attention to giving respect to HA in their services and give them regular encouragement to make them feel their own sense of meaning in leisure vacations and promote the formation of loyalty.
Finally, different measures can be used to enhance the attractiveness and competitiveness of the destination for the LA and HA. The recommendation willingness of HA is stronger, and RSHLVUE destination operators can be more biased to enhance their word-of-mouth marketing. Operators can give away brochures to them when they leave and ask them to promote the destination. The willingness to revisit is stronger among LA. Destination operators can establish a membership system to enhance their willingness to revisit through membership points and other membership tracking services, such as inviting members to participate in local festivals and events.

Limitations and Suggestions for Future Research
This study has some limitations, which can be addressed by future research. First, the data source of this study was only from Zhongyuan Township, China, potentially hindering the generalizability of the findings. Leisure vacation tourists from different geographical and cultural backgrounds may understand certain topics differently, and future research should involve different regions as well as different countries. Second, this study only uses age as a segmentation criterion and does not examine the variability of destination loyalty under other segmentation criteria. Factors such as marital status and health status may be considered subdivision criteria in subsequent studies. Finally, subsequent studies could attempt to introduce additional variables into the model to further explore more potential antecedents affecting RSHLVUE leisure vacation destinations' loyalty.

Data Availability Statement:
The data analyzed in this paper are proprietary, and therefore cannot be posted online.

Acknowledgments:
We are grateful to the questionnaire respondents and interviewers of this study for their help and support in obtaining the research data.

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

Appendix A. Detailed Analysis Process of Items of Tourist Well-Being
Appendix A.1. Coding Process of Semi-Structured Interview Data Semi-structured interviews were conducted in Zhongyuan Township, Jiangxi Province, China from 1-4 July 2021. Fifteen tourists were interviewed. The average interview time for each interviewee was about 45 min. To ensure that the interviewees understood these issues, the interviewers read the definitions and provided examples to explain the relevant attributes and definitions of "well-being" before each interview. The interviewers asked, "Did you have anything that affects your mood during summer health tourism in the countryside?", "What is your overall mood during summer health tourism?", etc.
More than 320,000 words of interview records were transcribed. Then, using grounded theory, interview data were coded to extract the dimensions and items. By using NVivo12 for open coding and main axis coding, the initial items and dimensions of the other part of the RSHLVUE tourist well-being scale were identified. First, the resident interview data numbered H1-H12 and the tourist interview data T1-T12 were coded. After cross-checking with literature and labels, the initial item elements were developed. Then, the initial item elements were classified and integrated to form the initial items. Finally, through the main axis coding, the initial items were conceptualized into the initial dimensions. The tourist interview data numbered T13-T15 and the resident interview data numbered H13-H15 were coded, and no new dimensions were found, which showed that the content analysis results were saturated. The specific process of the grounded theory is shown in Table A1. "The elderly tourists who come here mainly fancy the good air and comfortable summer climate. Because of the high forest coverage, this place is called a natural oxygen bar. The elderly who come here feel that their bodies are more comfortable than in the city, and their minds are clearer. In fact, they have taken a fancy to it." When elderly tourists feel that they are in good health, they show strong positive emotions, and their well-being increases steadily. As an elderly tourist (interviewee T8, woman, age 62) said: "I am not in good health. I have a headache when I use air conditioning in summer. It's cool in summer here, and I don't need to use air conditioning, which makes me feel comfortable. When I live here for a few days, I don't need to take some medicines for chronic diseases, and my body feels much more comfortable than in the city. Coming here in summer is my happiest time for health and wellness." There is a significant relationship between physical health and emotions. Health experience is a manifestation of the well-being of the elderly. Thus, health experience was an important dimension of the RSHLVUE tourist well-being scale.

Formation mechanisms of rural summer health leisure vacation destination loyalty
Information: Thank you for participating in this survey. We are doing a research project for the rural summer health leisure vacation for the urban elderly in the countryside. This questionnaire aims to understand the experiences of elderly tourists in rural summer health leisure vacation destinations. The questionnaire will require approximately 15 min to complete. Our questionnaire is anonymous, and your answers will be used in academic research only and will be kept absolutely confidential. Your support is very important for completing our research project and thesis, so thank you for your support and help. Section I. Main part of the questionnaire 33. I feel that this place has an important meaning to me (e.g., reunion of old friends, substantial improvement of health condition, some precious memories Below 3000 RMB; 3000-5000 RMB; 5000-7000 RMB; 7000-9000 RMB; 10000 RMB and above. 9. Compared with your peers around you, how do you think your physical health is? Very poor; Poor; General; Good; Very good. 10. Do you mind leaving contact information? It is convenient to ask you later. Thank you very much for your participation!