Roles of Tourism Involvement and Place Attachment in Determining Residents’ Attitudes Toward Industrial Heritage Tourism in a Resource-Exhausted City in China

Abstract

An in-depth discussion of place attachment in the relationship between residents’ perceived tourism impacts and their support for tourism is still lacking. The predictor of tourism involvement in this relationship has also been underestimated and little attention has been paid to industrial heritage tourism in relation to residents’ attitudes toward tourism development. To fill this gap, we extend upon the existing studies of residents’ attitudes toward tourism with place attachment (both place identity and place dependence) and tourism involvement based on social exchange theory, attitude theory, and the theory of planned behavior. A self-administered survey was completed by 336 residents of Huangshi, a city undergoing a transition to industrial heritage tourism in China. The findings show that residents’ support for tourism is the result of a complete behavior generation process. This has gradually formed through tourism involvement, cognition, affection, and behavior intention, emphasizing the importance of participation and affective attitude in determining residents’ attitudes toward tourism. To maintain the sustainable development of industrial heritage tourism in the economic transition from an old industrial region to new sectors, local authorities should attach more importance to strengthening residents’ native emotional bonds and concentrate on how to encourage local residents to participate in tourism activities.

1. Introduction

In the context of tourism, several studies have stressed that in tourist destinations, residents have a deeper perception and more profound understanding of urban changes than tourists, because they live, work and entertain in the destination every day [1]. Their support is considered a sign of success in sustainable tourism development [2]. With regard to residents’ sophisticated support behavior, social exchange theory (SET) has been widely used to shape how residents form supportive attitudes through perceived impacts (positive/negative or benefits/costs) on various tourism developments [3,4,5,6,7]. The theory describes the procedure of resident interactions as exchanges and evaluations of the benefits and costs. In the theory, if residents’ perceived positive impacts (benefits) outweigh their negative perceptions (costs), they will support tourism development; otherwise, resident support is likely to decrease [6,8].

Many scholars have explored predictors of residents’ perceptions of and support behavior (exchange procedure) toward sustainable tourism development, such as place identity [9]; place image and place attachment [1,10,11,12]; community attachment, community concern, and eco-centric attitude [13,14,15]; trust [2,16]; satisfaction and community commitment [17]; event attachment [18]; sense of place [19]; and empowerment [20].

However, little attention has been paid to tourism involvement, which is an attitude formed in the process of interactions with the social environment [21] and recreational activities [22,23,24]. Reid and Crompton [25] indicated that under the condition of low involvement, consistency between attitudes and behavior might not exist. Tourism involvement of residents was found to be closely related to the success of regional tourism planning [26] and the evaluation of tourism development [8]. The tourism involvement of residents plays an important role in accurately and verifiably evaluating attitudes [27] and understanding tourism behaviors [28]. Consequently, residents’ tourism involvement is likely to influence their exchange processes with tourism development, and ultimately influence their attitudes (perceived tourism impacts) and behaviors (support for tourism).

Besides the constructs that are related to the residents’ attitudes toward sustainable tourism development, why and how perceived tourism impacts act on support behavior also lack research attention. According to attitude theory, behavior involves cognitive and affective stages [29]. However, most of the studies on residents’ attitudes toward tourism have ignored the affective stage, only emphasizing the cognitive stage (perceived tourism impacts) [30]. As a core concept in environmental psychology, place attachment has been interpreted as involving the affective relationships between individuals and specific places [31,32]; in the context of tourism, place attachment has long been a criterion for assessing tourism impacts and stakeholders’ attitudes [10,12,33,34]. However, in terms of attitude theory, how the positive (maybe) perceived tourism impacts understood by residents (cognitive stage) influence their attachment level to the place (affective stage) has yet to be accurately and directly examined.

Studies have focused little on industrial heritage tourism (IHT) in these contexts. For resource-exhausted cities, the development of IHT may be a good method to enhance urban revitalization and reconstruction in the economic transition periods from an old industrial region to new sectors (e.g., Ruhr in Germany, North-eastern England and Wales) [35], especially for the promotion of the local economy and the sustainability of the regional industrial culture [36,37,38,39]. Although this momentous transition (in tourism) is mainly formulated by the government and decision makers, the role of residents is rarely mentioned. In similar cities from developing countries, the issues of resource-exhaustion and tourism transition have not been particularly prominent until now. Residents’ attitudes toward and support for IHT are critical factors in its sustainable development and effective management in the economic transition period from an old industrial region to new sectors [40].

Based on the above background, the purpose of this study was to use the SET [8], the theory of planned behavior [41], and attitude theory [29] to examine the relationships between tourism involvement, perceived tourism impacts (two parts of cognitive attitude), place attachment (newly introduced affective attitudes), and support (behavior) in a resource-exhausted city (refers to those cities where the development of mineral resources has entered a period of decline or exhaustion, and their accumulated mining reserves have reached more than 70% of the total determined [42]) in China that is currently transitioning to IHT. We further explored whether place attachment can be switched to a non-prepositive construct that guides the influence of attitude (both involvement and perceived impacts) on behavior (support).

We provide the following theoretical contributions. Firstly, this study is a response to calls for exploring the affective stages in the existing studies of residents attitudes toward tourism based on SET theory [2,30,43,44]. As a construct for expressing affective attitudes, place attachment has yet to be examined in the relationship between residents’ perceived impacts and their support behaviors in the context of tourism. Secondly, the authentic and more subjective construct, tourism involvement, is used as the predictor, which has been neglected in previous studies. Those studies ignored the fact that the more an individual is involved in tourism activity, the more concrete their cognitive and affective attitudes toward tourism development [25,27,28]. Thirdly, our findings help to determine how and why place attachment influences the formation of residents’ support for tourism after the cognitive stage of perceived tourism impacts. Finally, as a response to the calls for more attention to be placed on a special aspect of tourism, industrial heritage tourism [40], we conducted a quantitative analysis and developed managerial implications for resource-exhausted cities in developing countries as a potential reference for the early stages of the transition to IHT.

We outline some policy recommendations on how the residents’ tourism involvement influences their perception of tourism impacts and their affective bonds to their living place. The findings will help local authorities and destination managers to formulate plans for the sustainable development of IHT using cognitive (perceived tourism impacts) and affective (place attachment) attitudes.

2. Literature Review

2.1. Residents’ Perceived Tourism Impacts and Their Support for Tourism

Guided by the SET, studies have thoroughly examined the relationships between different kinds of residents’ perceived tourism impacts and their support for tourism [1,3,4,13,18,45,46,47]. As the core content and structure of relevant studies, when residents perceive more positive (benefits) than negative impacts (costs) in the context of tourism development, they are more likely to support it [8]. Despite the suggestion by Stylidis et al. [1] to use the non-forced scale (a bipolar scale ranging from strongly negative to strongly positive) to show resident’s real perception of tourism impacts, this method would only provide an overview of residents’ evaluations and neglect some impacts, (e.g., a mean value of the bipolar scale of 3.5 denotes positive evaluations from most respondents, but we cannot say that the negative part does not exist, and this part may have been neglected in the process of improving residents’ attitudes toward tourism development). Considering the more popular traditional costs and benefits measurement, we continued to use the perceived positive impacts to interpret the perceived benefits and negative impacts to interpret the costs.

Similar to the concept of the triple-bottom-line [48], perceptions of tourism impacts are usually multidimensional, including perceptions of economic, socio-cultural, and environmental impacts, to ensure that the evaluation of the sustainability of tourism development is effective. Living standard improvements, more job opportunities, and infrastructure construction are advantages for residents that emerge from the development of tourism [1,12]. Through the development and spread of traditional culture in tourism planning, residents’ awareness of cultural heritage preservation is promoted, and the community-to-community exchanges are enhanced [44,49]. Tourism activities encourage residents to participate in the local environment, deepening their awareness of environmental protection [18]. The development of tourism may also create economic, socio-cultural, and environmental issues, such as traffic problems, tax burdens, and crime [50]; crowds, noise/light pollution, and prostitution [30]; the overuse of local heritage, and even conflicts between visitors and residents [19]. To summarize, previous studies have shown that the more positively economic, socio-cultural, and environmental impacts are perceived by local residents, the more willing they are to support the development of tourism. Conversely, if they perceive more negative impacts from tourism, their support will decrease [4,7,11,16,18,45].

The relationships and impacts mentioned above have been applied to various forms of tourism development, such as mass and alternative tourism [5], mega-events [16], festivals [44], and heritage tourism [19]. However, in the context of IHT, studies are limited and have mainly used qualitative methods [37,39,51]. Otgaar et al. [52] stated that the development of industrial tourism (including IHT) could contribute to the income and image of the host cities. Under some conditions, interests can converge through developing industrial tourism between cities and tourism companies. Hashimoto and Telfer [53] suggested that the development of IHT can create many positive impacts (such as providing opportunities for local communities to learn about the past, promoting the transformation of derelict buildings, and revitalizing community development, but the key point is to determine which buildings should be conserved and what stories about the past should be prepared for local residents. Vargas et al. [40] showed that residents’ perceived impacts of industry tourism (the indicators were borrowed from studies about ski tourism [54], island tourism [55], and coastal tourism [56]) influence their support for it. They stated that the effective management of that influence would only appear in the early stages of tourism development, especially in industry tourism. Consequently, for a resource-exhausted city in the early transitional state of IHT, exploring residents’ perceptions of tourism impacts and their support for this kind of tourism is particularly necessary. Based on the SET and aforementioned discussion, we propose the following hypotheses:

Hypothesis 1 (H1).

Residents’ perceived positive impacts of industrial heritage tourism positively influence their support for it.

Hypothesis 2 (H2).

Residents’ perceived negative impacts of industrial heritage tourism negatively influence their support for it.

2.2. Place Attachment

Applied to tourism studies in the 1980s [57], place attachment is one of the most popular concepts that describes the affective relationships between individuals and places in environmental psychology. It is defined as the result of an individual’s accumulated social and physical experiences in a specific place or environment [10], or the emotional bond between an individual and a particular environmental setting [58]. Although debate is ongoing about the use of place attachment and other concepts that also define the human-land relationship, such as sense of place [59], community attachment [3], and place bonding [60], place attachment seems to be the mainstream concept that is most frequently used [61]. In the current study, place attachment was considered to be more suitable for demonstrating local residents’ attributes of life than other concepts [10,62], and it is the non-economic factor that was used to interpret their attitudes to tourism development [63].

Based on the theory of attitude, attitude is composed of cognition, affection, and behavior. Among these factors, the position of affection is in the middle [64]. However, in most studies on residents’ support for tourism, in line with SET, residents’ perceived tourism impacts (cognition) and their support for tourism (behavior) have been the only factors considered, and affection content, which describes how the behavior is formed through cognition, is omitted. To fill this gap, some studies have attempted to use affective constructs. Song et al. [44] used “affective impacts” as the affection part in determining residents’ attitudes toward tourism, but these affective impacts were parallel to the positive and negative perceived festival impacts. Vong et al. [30] used “sense of place” as the affective construct in Macao, China, and showed that sense of place not only played an affective role, but also mediated the relationship between residents’ perceived impacts of a casino and their support for it, by attempting to improve residents’ attitudes toward tourism. Consequently, although place attachment is usually used as a predictor [10,12,33,34], considering its emotional bonds and residents’ life attributes, we applied place attachment to affection’s position between cognition and behavior in the context of residents’ attitudes toward tourism development.

In the context of tourism, place attachment usually has two dimensions: place dependence and place identity [21,28,58,65,66,67,68]. Place dependence refers to the functional bond that relates to how individuals evaluate the settings, activities, or attractions in specific places [69,70]. It emphasizes the irreplaceability of the places where individuals live, work, and rest. Place identity is usually defined as the emotional bond that an individual feels with a place [69], and the linkage formed through self-regulation processes [66]. This identity has been shown and confirmed by individuals in the context of tourism [67], and leads to the belief that they belong to the place [30]. The two dimensions of place attachment, the scales of Williams and Vaske [68] and Kyle et al. [67], are leading examples. The difference between the scales is that Williams and Vaske’s operational indicators are more detailed in expressing place identity and place dependence. For instance, in the dimension of place identity, the scale of the former explores whether the place could be felt like part of the respondent or reflect the respondent. Owing to its fewer indicators, the reliability and validity of the scale by Kyle et al. [67] is usually more acceptable in the tourism field [21,34,64,70].

To date, few studies have examined whether place dependence and place identity can influence residents’ support for tourism. Buonincontri et al. [71] showed that visitors with a high place attachment to heritage sites can positively influence sustainable heritage behavior (e.g., lend support to the protection of heritage sites with money). Lee [47] suggested that locals who are highly tied to a place are likely to have more interest in tourism sustainability. If residents feel consistency between tourism development and place identity, they will be more likely to agree with tourism [9]. Schuster et al. [72] studied residents’ attitudes towards natural and heritage tourism through place attachment in three communities near the Hudson River in America (Beacon, Cold Spring, and Kingston). They found that significant differences existed among communities in terms of place dependence and place identity, and residents with strong identification with their communities were more willing to support the development of tourism. Only Kingston’s residents showed a positive relationship between their dependence and support for tourism. Hence, the tighter the resident’s bonds to the functions and emotions of a place, the more likely they are to favor tourism development. Therefore, the following hypotheses are presented:

Hypothesis 3 (H3).

Residents’ place identity positively influences their support for tourism development.

Hypothesis 4 (H4).

Residents’ place dependence positively influences their support for tourism development.

To date, few studies have provided direct evidence to support the influence of residents’ perceived tourism impacts on their attachment to a place. However, some studies may have indirectly provided some evidence for the relationship. Wang and Xu [9] stated that when residents’ needs and preferences regarding a city’s physical, social-cultural, and environmental resources are satisfied, their identity and attachment to the place increase. Gu and Ryan [73] found that the perception of preservation emerged from the sense of being proud at Beijing Hutong, which led to a specific identity tied to the place. This shows that the more positively residents perceive the local environment, the stronger the bonds they attach to the place [74]. Ouyang et al. [2] showed that the more positively residents perceive tourism impacts, the more positive their emotions; conversely, a negative perception of tourism leads to negative emotions. Hosany et al. [75] deemed that positive and negative emotions influence the extent to which individuals identify with or depend on a place, proving that emotions are predictors of place identity and place dependence, respectively (not just place attachment). Therefore, a relationship appears between residents’ perceived tourism impacts and their place identity and dependence. Residents’ positive perceptions of IHT impacts may lead them to strongly identify with and depend on the place in which they live. In contrast, the perception of more negative impacts of IHT will lead to worse evaluations of and feelings about a place. Consequently, we formulated the following hypotheses:

Hypothesis 5 (H5).

Residents’ perceived positive impacts of industrial heritage tourism positively influence their place identity.

Hypothesis 6 (H6).

Residents’ perceived positive impacts of industrial heritage tourism positively influence their place dependence.

Hypothesis 7 (H7).

Residents’ perceived negative impacts of industrial heritage tourism negatively influence their place identity.

Hypothesis 8 (H8).

Residents’ perceived negative impacts of industrial heritage tourism negatively influence their place dependence.

2.3. Tourism Involvement

Involvement is a complex concept that is related to an individual’s attitudes and decisions in marketing [76]. If an object or situation is personally relevant to an individual, they are likely to be involved in it [77]. With different levels of involvement, individuals will differ in decision making, thereby motivating their behaviors [78]. Therefore, involvement could be a predictor of an individual’s behavior [79]. In the context of tourism, the concept of involvement is usually defined as motivation, arousal, and interest in specific tourism activities or products, which could be triggered by stimuli and situations [22,23,24]. An individual’s tourism involvement can interact with the social environments of the surrounding places [21], forming emotional bonds to the place [80], or the perception of a destination experience to form [79].

Debate exists regarding the multidimensional constructs of involvement [21,81], such as the Personal Involvement Inventory proposed by Zaichkowsky [77] and the Consumer Involvement Profile introduced by Laurent and Kapferer [78]. Due to the different target objects, multidimensional constructs of involvement may not be balanced in reflecting the contents and effects (in each dimension) [23]. As some of the indicators could not meet the requirements of the measurement model, Prayag and Ryan [21] failed to use the multidimensional scale of the Consumer Involvement Profile (CIP) to predict tourists’ loyalty model in Mauritius. However, according to the results of confirmatory factor analysis (CFA), a unidimensional construct consisting of six indicators, which were initially used to describe the importance, pleasure, and sign (from the CIP scale), was found to significant influence tourists’ loyalty. Lu et al. [76] only chose three indicators to express people’s perceived pleasure in participating in activities to build a unidimensional construct of involvement employing the hedonic and pleasure scales by Funk et al. [81], and explored the significance of the relationship between image and satisfaction. Based on the above studies, to reduce the complexity of our structural model and avoid the potential negative effects of multidimensional constructs of involvement, we chose a unidimensional construct for this study.

Tourism involvement is a crucial determinant of sustainable tourism. When residents are stimulated by the newly developed local tourism industry, interest is aroused in relation to tourism participation. Then, they are likely to join in and perceive benefits from tourism development, such as the increase in employment opportunities and the preservation of local culture. Therefore, the support for tourism is also expedited [47]. An association between the degree of involvement and perceived negative impacts has also been examined [82]. The more an individual participates, the better they understand some of the costs of tourism development [83]. Residents will better grasp the actual urban tourism development situation. Therefore, they will know whether they are interested in local tourism activities and support there development through comparing the benefits and costs [84]. Hence, we present three additional hypotheses:

Hypothesis 9 (H9).

Residents’ tourism involvement positively influences their perceived positive impacts of industrial heritage tourism.

Hypothesis 10 (H10).

Residents’ tourism involvement negatively influences their perceived negative impacts of industrial heritage tourism.

Hypothesis 11 (H11).

Residents’ tourism involvement positively influences their support for industrial heritage tourism.

The relationship between tourism involvement and place attachment has been well examined in prior studies. Some used multidimensional constructs of involvement. Kyle et al. [24] deemed that an individual’s attraction involvement is related to place identity, whereas self-expression involvement influences place dependence. Using the same theoretical model, Gross and Brown [85] found that centrality and food/wine involvement positively influenced both place identity and place dependence, and attraction involvement was shown to negatively influence place dependence. Some studies used unidimensional constructs of involvement and place attachment to show that tourism involvement is a statistically significant predictor of place attachment [21,28]. However, few studies have examined the relationship between the unidimensional involvement and multidimensional place attachment of residents. When residents’ interest, arousal, motivation, and participation in local tourism increase (different local experience), they will have a deeper understanding of and connection to their local area, which influences their human-land emotion and local function cognition. To minimize the research gap and examine the connections, the two following hypotheses are presented:

Hypothesis 12 (H12).

Residents’ tourism involvement positively influences their place identity.

Hypothesis 13 (H13).

Residents’ tourism involvement positively influences their place dependence.

2.4. Proposed Framework

Based on the above discussion, our proposed framework (Figure 1) considers that residents’ support for the development of IHT is affected by their evaluation of cognitively perceived tourism impacts and affective place attachment. The degree of resident involvement in tourism directly influences their support for the development of IHT.

3. Research Method

3.1. Study Location

Huangshi (Southeastern Hubei province, China) was chosen as the focus for the current study for two reasons. The first reason is that few studies have explored the early stage of development in tourism destinations [1]. In the early stage of tourism development, residents usually have high hopes or expectations about tourism impacts on future tourism development, especially when the host city is experiencing economic transition [54]. Therefore, during this period, their opinions are conducive to the proposal of pertinence measures and the sustainable development of tourism because any improvement aimed at their expected perception of tourism impacts will let them realize the benefits obtained from IHT development. The other reason is that prior studies failed to test residents’ attitudes toward tourism in the context of IHT. More specifically, in developing countries, the attitudes of residents in resource-exhausted cities toward tourism development provide crucial evidence for whether industrial transition tourism will be successful and sustainable.

Huangshi is a prefecture-level city next to Wuhan (the capital of Hubei Province), with a population of 2.68 million (as of 2017) [86]. An important resource-based city with rich iron ore and non-ferrous metals, Huangshi is one of the birthplaces of the Chinese Bronze Culture and the cradle of the national industry in modern China [87]. It supplied most of the raw materials to Hanyang Iron Works, which was Asia’s largest and earliest steel conglomerate in the Qing Dynasty in the 1910s. After the liberation of China, the city played the role of the country’s important industrial raw materials base until the 1990s. As resources were gradually exhausted, the drawbacks of its heavy dependence on industry began to appear, and urban economic development entered a stagnant period. In 2009, Huangshi was defined by the State Council as a resource-exhausted city, and industrial transition became the primary task of development [88]. In the process of transition, some industrial enterprises moved out of the urban area or stopped production because of benefit problems, and many industrial sites appeared. Therefore, the development of IHT became an important direction for the local government when it attempted to conduct industrial transition. More recently, local governments have successively renovated and protected these industrial sites, and reactivated them through tourism and leisure development. In 2018, the National Mining Park (traced back to 1890, known as the largest open pit in Asia), Daye Ironworks Former Site (traced back to 1913, part of Asia’s largest and earliest steel conglomerate), and the Huaxin Cement Former Site Museum (traced back to 1907, the cradle of China’s cement industry), were chosen in the first batch to be included in the Industrial Heritage Protection List in China [89]. Huangshi has also organized some high-quality events related to mining and metallurgical culture, such as the International Mining and Metallurgical Culture Tourism Festival and the Forum and Symposium on Mining and Metallurgical Culture, to improve its influence in the field of IHT. After brief initial development, the number of tourists rose rapidly from 7.91 million in 2010 (initial time hosting related events) to 22.20 million in 2017 [90,91]. This shows that the transition to IHT has achieved positive results. In the future, Huangshi plans to become the most beautiful industrial heritage tourism city in China, and attain the status of a World Heritage Site. IHT will be an important booster of urban sustainable development in the future.

3.2. Questionnaire Development

Based on the proposals of DeVellis [92] and Churchill [93] for scale development and construct management, a survey instrument was developed to examine residents’ involvement, perceptions of tourism impacts, place attachment, and support for industrial heritage tourism. The questionnaire was composed of seven main sections described below.

The tourism involvement section was adopted from Prayag and Ryan [21] to measure the resident’s motivation for, arousal to, and interest in IHT. The indicators are whether IHT is pleasant, whether the residents attach importance to IHT, whether the residents have interest in IHT, whether the travel to IHT is similar to giving a gift to residents, whether residents obtain pleasure by interacting with various IHT sites, and whether residents can tell much about a person/family through whether or not they travel around Huangshi.

The perceived tourism impacts section contains 11 indicators adopted from Prayag et al. [7]. This scale has been well examined in studies on residents’ support for tourism [2,16,18]. Among the indicators, six indicators constitute the perceived positive tourism impacts, measuring residents’ assessments of some positive impacts caused by IHT development. The indicators discuss a closer community, improvement of the city’s image, tourist destination promoting, environmental conservation and protectionism improvement, environmental awareness, and more employment opportunities. Similarly, perceived negative impacts are assessed by the other five indicators to measure the residents’ perceptions of some negative impacts, including littering and disorganization, natural environment damage, pollution, cost of living, and the crime rate.

The place attachment section contains eight indicators adopted from Kyle et al. [67], and its validity and reliability have been shown in previous tourism studies [34,65,94]. Place identity and place dependence are the two dimensions of place attachment. Indicators from place identity refer to the emotional bonds to the place, including “Huangshi means a lot to me”, “I am very attached to Huangshi”, “I identify strongly with Huangshi”, and “Huangshi is a very special destination to me”. Indicators from place dependence refer to the functional bonds to the place, including “I enjoy visiting Huangshi more than any other place”, “I obtain more satisfaction out of visiting Huangshi than any other place”, “Visiting Huangshi is more important to me than visiting any other place”, and “I would not substitute any other place for the type of experience I have in Huangshi”.

The residents’ support for tourism development section was adopted from Stylidis et al. [1], which has been well examined in tourism studies [10,11,12]. The respondents were asked to indicate whether they will support the development of IHT. The indicators are “public finance for IHT promotion”, “further IHT development”, and “increase in the number of tourists”.

All these indicators were measured on a five-point Likert scale ranging from 1 = strongly disagree to 5 = strongly agree. Finally, some social demographic factors were measured using seven indicators adopted from Stylidis et al. [1], including sex, age, education, income, and length of residence.

According to Podsakoff et al. [95], when hypothesized relationships between constructs are studied, a systematic effect bias may exist, especially in behavioral studies, namely common method variance (CMV). To minimize the potential CMV impact, the questionnaire was designed to follow the guidelines of Podsakoff et al. [96]. Hosany et al. [75] suggested that Harman’s single factor test is an effective method for evaluating the influence of CMV in the field of tourism study. CMV exists when the results of the exploratory factor analysis (all the constructs together with non-rotation) have just one single factor, or when more than one factor is shown but the total variance of any factor accounts for more than 50% [97]. In this study, six factors with eigenvalues higher than 1 were proven by the exploratory factor analysis (EFA), and the first factor accounted for 30.40% (less than 50%) of the total variance (out of 71.28%). This demonstrates that the data do not contain CMV bias.

3.3. Sampling and Data Collection

To guarantee the quality of the sampling process, a pilot study and a main study were conducted. Because the indicators originated from English-speaking countries, the initial self-administered questionnaire had to be translated into Chinese. A lecturer who majored in English at Henan University was invited to translate the questionnaire. Based on the back translation method [98], after the first Chinese edition was completed, another two professors in the field of tourism in Henan University were also invited to check whether there were inexplicit expressions from the questionnaire in the context of tourism terminology. Then, the second edition was prepared for the pilot study on 10 January 2019. We randomly chose 20 respondents living next to the Huaxin Cement Former Site Museum to fill in the self-administered questionnaires. Based on their feedback about the comprehension, clarity, and length [44], the questionnaire was revised again, and the final edition was employed for the main study.

To minimize potential bias, the main study involved the use of systematic random sampling, in accordance with Tournois and Djeric [12] and Andriotis [99]. The target respondents had to be permanent residents and above 18 years of age. To guarantee the randomness of samples, questionnaires were collected both in main streets with large streams of people (one respondent per 10 people and per household) and from representative IHT sites in Huangshi (even in suburban districts) to cover urban and suburban areas. The main study was conducted from 24 January 2019 to 18 February 2019. As the Spring Festival is celebrated from 4 February to 10 February, and the traditional festival lasted for 15 days, the sampling period was inside the Spring Festival tourism season. Usually, the total number of local tourists increases during the Spring Festival holiday. Therefore, this tourism season provided convenience for the sampling process. During the in-person survey, the purpose of the survey was expressed to the respondents first, and they were then asked if they were willing to participate. In total, 379 questionnaires were collected, and 43 samples were removed due to the excessive number of incomplete sections, so 336 samples were retained. When using the structural equation modeling (SEM), a sample size of more than 200 is usually required, or each indicator should match at least 10 samples [100,101]. As 336 (>200) samples were collected and with 27 overall indicators (336 > 27 × 10 = 270), the retained samples were sufficient for subsequent data analysis.

3.4. Data Analysis

Data were analyzed in several steps using Statistical Product and Service Solutions 22.0 (SPSS) and AMOS 22.0 (an added SPSS module). SPSS 22.0 was used to (1) determine the demographic profiles of the respondents by frequency analysis, (2) measure the reliability of each construct (internal consistency) by Cronbach’s alpha analysis, (3) generate the mean score for each construct using descriptive analysis, and (4) examine the CMV by exploratory factor analysis (EFA) [95]. Following the two-step SEM approach [102], AMOS 22.0 was used to test the model fit of the measurement model and structural model. Using the maximum likelihood method, the reliability, convergent validity, and discriminant validity were examined in the confirmatory factor analysis (CFA). Then, the relationships between constructs were shown by the standardized coefficients (95% confidence interval).

4. Findings

4.1. Demographic Profile

The demographic characteristics of the respondents in this study are shown in

Table 1. Profile of respondents (N = 336).

Sample			Huangshi Census 2010 a		Hubei 1% Sample Census 2015 b
Sex	No.	Ratio		Ratio		Ratio
Male	172	51.2%	Male	51.84%	Male	51.00%
Female	164	48.8%	Female	48.16%	Female	49.00%
Age group, years
18–24	48	14.3%	0–14 c	17.39%	0–14	15.03%
25–34	76	22.6%	15–64	74.81%	15–64	73.93%
35–44	66	19.6%
45–54	59	17.6%
55–64	54	16.1%
65 and above	33	9.8%	65 and above	7.80%	65 and above	11.04%
Monthly income (RMB)			Statistical Bulletin of National Economic and Social Development for Huangshi (2017) d Average annual income (RMB) Rural residents: 13,973 Urban residents: 32,535
Less than 1400	23	6.8%
1401–3000	127	37.8%
3001–4000	69	20.5%
4001–5000	43	12.8%
5001–6000	25	7.4%
6001–7000	10	3.0%
More than 7001	39	11.6%
Education
Junior school or below	47	14.0%	75.74%		70.68%
High school	158	47.0%	16.46%		17.85%
College/University	113	33.6%	7.80%		11.47%
Postgraduate	18	5.4%	Included in College/University
Length of residence (years)
Less than 1 year	21	6.3%	- e
1–10	53	15.8%	-
11–20	29	8.6%	-
More than 20 years	233	69.3%	-

Note: ^a Source: Huangshi Bureau of Statistics [103]; ^b Source: Hubei Province Bureau of Statistics of China [104]; ^c The age groups in the census in China are usually divided into 0–14, 15–64 (15–59/60–64), and 65 and above by the National Bureau of Statistics of China [105]; ^d Source: Huangshi Bureau of Statistics [91]; 1 RMB = USD $0.1485 (as of 30 April 2019); ^e Data lacking for direct comparison.

. Comparing such data with national or regional census data is useful for testing the representativeness of the sample [1,30]. We used the National Census of the China (conducted every 10 years) and the 1% National Sample Census (conducted every five years) as references. Similar to the two censuses, 51.2% of the respondents were male and 48.8% were female. Respondents aged between 25 and 34 years (22.6%) were the largest group, followed by the ages of 35–44. More than one-third of the respondents had a monthly income of ¥1401–3000 and 64.3% had finished senior school or obtained higher degrees. Nearly 70% of the respondents had been living in Huangshi for more than 20 years. Appendix A provides the descriptive statistics (means, standard deviations, skewness, and kurtosis) for each construct. The mean value of support (over 4.00) suggests that most of the residents agreed to support the development of industrial heritage tourism, and they identified both the positive (3.82) and negative (2.23) tourism impacts well. Their scores for involvement in tourism and attachment to the place were both above 3.44, indicating relatively strong agreement with their participation and emotion.

4.2. Measurement Model

Before applying the measurement model, the univariate skewness, kurtosis, and Mardia’s coefficient were examined (Appendix A) to ensure that the normality assumption would not be violated. Although the skewness (<3) and kurtosis (<10) were consistent with the recommendation [106], the multivariate kurtosis (t = 38.668 > 5) indicated a multivariate non-normal distribution [107]. Therefore, the AMOS with its 10,000 bootstraps method, was used to generate an approximate standard error for estimating the measurement model and structural model, where the assumption of multivariate normality needs not be satisfied [44,108]. When using the bootstrapping approach for determining the parameters’ standardized errors in SEM, it is necessary to fix the indicator path rather than the factor variance; otherwise, the standardized errors will be inflated [109]. Specifically, following the guidelines of bootstrapping for addressing non-normal data by Byrne [107] and Bollen and Stine [110], the procedures of the maximum likelihood method and 95% bias-corrected confidence intervals were chosen.

The initial results indicated that one indicator (increased the cost of living) from the perceived negative impacts had a low standardized factor loading below 0.5. After elimination, the external model fit was acceptable with χ²(309) = 541.831 (p < 0.05), χ²/df = 1.753, Comparative Fit Index (CFI) = 0.960, Non-normed-fit Index (NNFI) = 0.954, and Root Mean Square Error of Approximation (RMSEA) = 0.047 [101,106,111]. As shown in

Table 2. The confirmatory factor analysis of the measurement model. IHT: industrial heritage tourism.

Constructs/Indicators	Factor Loading	S.E. Boot	Cronbach’s α
Involvement (INVL)			0.915
I receive pleasure from IHT here	0.724
I attach great importance to IHT in Huangshi	0.831	0.067
I have considerable interest in Huangshi as an IHT destination	0.852	0.066
Local travel or leisure at industrial heritage sites is a bit like giving a gift to one’s self	0.811	0.072
I give myself pleasure by becoming involved in the various industrial heritage sites to see here	0.804	0.076
I can tell much about a person/family by whether or not they travel around Huangshi	0.788	0.070
Perceived Positive Impacts (PPI)			0.919
The development of IHT could have positive impacts by…
Bringing the local community closer	0.734	-
Improving Huangshi’s holistic image	0.864	0.069
Promoting Huangshi as a tourist destination	0.836	0.074
Improving environmental conservation and protectionism	0.787	0.098
Raising environmental awareness	0.853	0.089
Providing locals with employment opportunities	0.791	0.085
Perceived Negative Impacts (PNI)			0.850
The development of IHT could have negative impacts by…
Increasing littering and disorganization in the city	0.781	-
Damaging the natural environment	0.780	0.069
Increasing noise, air, and visual pollution	0.763	0.081
Increasing crime	0.742	0.082
Place Identity (PI)			0.869
Huangshi means a lot to me	0.837
I am very attached to Huangshi	0.856	0.044
I identify strongly with Huangshi	0.743	0.074
Huangshi is a very special destination to me	0.727	0.066
Place Dependence (PD)			0.854
I enjoy visiting Huangshi more than any other place	0.881	-
I obtain more satisfaction out of visiting Huangshi than any other place	0.606	0.060
Visiting Huangshi is more important to me than visiting any other place	0.841	0.056
I would not substitute any other place for the type of experience I have in Huangshi	0.764	0.060
Support. for tourism development (SUP)			0.923
I support…
Public finance for IHT promotion	0.877	-
Further IHT development	0.928	0.050
An increase in the number of tourists	0.884	0.049

Note: S.E. boot: the standardized errors (S.E.) were estimated by the bootstrapping maximum likelihood approach with 10,000 resamples in AMOS; the factor loadings of all indicators were significant (p < 0.001) as determined by the two-tailed 95% bias-corrected percentile method; the Cronbach’s α values of all the constructs exceeded 0.757, indicating a good reliability [115].

, except for the second indicator of place dependence (0.606), all the standardized factor loadings for each construct’s indicator exceeded 0.7 (ranging from 0.724 to 0.928), which indicated the indicator validity [112]. The internal consistency was examined through determining the composite reliability of each construct. The values were all above the recommended cut-off criterion of 0.6 (ranging from 0.851 to 0.925;

Table 3. The discriminant validity.

Construct	INVL	PPI	PNI	PI	PD	SUP
INVL a	0.644 b	0.092 d (0.304 c)	0.019 (−0.138)	0.091 (0.301)	0.141 (0.376)	0.075 (0.273)
PPI	0.059 e	0.660	0.131 (−0.362)	0.198 (0.445)	0.104 (0.322)	0.314 (0.560)
PNI	0.063	0.060	0.588	0.038 (−0.195)	0.056 (−0.237)	0.133 (−0.365)
PI	0.059	0.050	0.063	0.629	0.299 (0.547)	0.137 (0.370)
PD	0.060	0.059	0.065	0.053	0.609	0.071 (0.266)
SUP	0.057	0.054	0.055	0.054	0.060	0.804
Composite Reliability	0.916	0.921	0.851	0.871	0.859	0.925

Note: ^a INVL = Involvement, PPI = Perceived Positive Impacts, PNI = Perceived Negative Impacts, PI = Place Identity, PD = Place Dependence, SUP = Supports for Tourism Development; ^b Average variance extracted; ^c Correlations between pairs of constructs; ^d Inter-construct squared correlation values; ^e Standardized errors between pairs of constructs (estimated by the bootstrapping maximum likelihood approach with 10,000 resamples in AMOS).

) [113]. Additionally, the average variance extracted (AVE) values of each construct were also higher than 0.5 (ranging from 0.588 to 0.804), which confirmed the scale’s convergent validity [102]. The AVE values (along the diagonal) were found to be higher than the inter-construct squared correlation values (above the diagonal, outside the parenthesis), suggesting a tenable discriminant validity [114].

4.3. Structural Model

The structural model was examined with the bootstrapping maximum likelihood approach based on the validated measurement model. The results revealed a good fit for the structural model with χ²(311) = 622.802 (p < 0.05), χ²/df = 2.003, CFI = 0.946, NNFI (TLI) = 0.939, and RMSEA = 0.055. This shows that a good fit between the proposed theoretical model and the empirical data. The structural model explained 33.7%, 43.5%, and 11.5% of the variance in residents’ support for IHT, place attachment, and perceived tourism impacts, respectively. As shown in

Table 4. Hypothesis testing.

Hypothesis	Path	Estimate	S.E. Boot	p	Testing Results
H1	perceived positive impacts → support	0.435	0.093	0.000	Supported
H2	perceived negative impacts → support	−0.192	0.067	0.000	Supported
H3	place identity → support	0.121	0.063	0.043	Supported
H4	place dependence → support	−0.001	0.056	0.845	Rejected
H5	perceived positive impacts → place identity	0.382	0.078	0.000	Supported
H6	perceived positive impacts → place dependence	0.202	0.086	0.002	Supported
H7	perceived negative impacts → place identity	−0.049	0.091	0.428	Rejected
H8	perceived negative impacts → place dependence	−0.127	0.103	0.031	Supported
H9	involvement → perceived positive impacts	0.307	0.066	0.000	Supported
H10	involvement → perceived negative impacts	−0.146	0.058	0.025	Supported
H11	involvement →support	0.087	0.062	0.076	Rejected
H12	involvement → place identity	0.187	0.074	0.005	Supported
H13	involvement → place dependence	0.307	0.090	0.000	Supported

Note: Estimate: the standardized β values; S.E. boot: the standardized errors were estimated by the bootstrapping maximum likelihood approach with 10,000 resamples in AMOS; p: a significance value determined by the two-tailed 95% bias-corrected percentile method.

, 10 hypotheses were supported, whereas H4, H7, and H11 were not. In the relationship between residents’ perceived impacts and their support for tourism, both residents’ perceived positive and negative impacts directly and significantly influenced their support for tourism, supporting H1 (β = 0.435, standardized errors estimated by bootstrapping approach (S.E. boot) = 0.093, p < 0.001) and H2 (β = −0.192, S.E. boot = 0.067, p < 0.001). In the relationship between residents’ perceived impacts and place attachment, residents’ perceived positive impacts directly, and significantly influenced place identity and place dependence, but residents’ perceived negative impacts only significantly influenced place dependence, providing support for H5 (β = 0.382, S.E. boot = 0.078, p < 0.001), H6 (β = 0.202, S.E. boot = 0.086, p < 0.01), and H8 (β = −0.127, S.E. boot = 0.103, p < 0.05), with H7 (β = −0.049, S.E. boot = 0.091, p > 0.05) being rejected. The findings also supported H3 (β = 0.121, S.E. boot = 0.063, p < 0.05), confirming that place identity was positively and significantly related to residents’ support for tourism. Conversely, place dependence did not have a significant influence on this support, leading to the rejection of H4 (β = −0.001, S.E. boot = 0.056, p > 0.05). Finally, the relationships between residents’ tourism involvement, perceived tourism impacts, and place attachment were significant, supporting H9 (β = 0.307, S.E. boot = 0.066, p < 0.001), H10 (β = −0.146, S.E. boot = 0.058, p < 0.05), H12 (β = 0.187, S.E. boot = 0.074, p < 0.01), and H13 (β = 0.307, S.E. boot = 0.090, p < 0.001). However, H11 (β = 0.087, S.E. boot = 0.062, p > 0.05) was rejected, since the influence of residents’ tourism involvement on their support for tourism was not significant.

5. Conclusions

In this study, we explored a conceptual and integrative framework of residents’ support for tourism development based on SET following the guidelines of attitude theory. The main differences in this study from what has been previously suggested are as follows. Firstly, place attachment was adopted as a new affective response (affective attitudes) instead of emotion or sense of place in other studies. Secondly, place attachment, which used to be a predictor in prior studies of residents’ support for tourism, was also set as the endogenous construct to explore how it influenced residents’ support for tourism after the cognitive attitude of perceived tourism impacts. Thirdly, we chose a unidimensional construct of tourism involvement instead of a multidimensional construct to examine its influence on residents’ cognitive and affective attitudes and support behaviors. Fourthly, different from prior studies that focused on various general forms of tourism development (e.g., history tourism, mega-events, and gambling) [1,16,30,45], we targeted IHT in a developing country’s resource-exhausted city. As such, based on the newly introduced content and relationships, we expanded the boundaries of the research on residents’ support for tourism development.

The findings are consistent with prior studies where residents’ perceived impacts (positive and negative) were shown to significantly influence their support for tourism [1,3,18,47]. The results confirmed the SET: if residents perceive greater benefits from IHT development, they are more likely to support its sustainable development in the future; conversely, if they feel that a few negative factors have resulted from the development of IHT, they will not have a preference for it [8]. In terms of local residents, the results also provide support for the effective transition to IHT in a developing country’s resource-exhausted city, providing a realistic reference for similar cities during the economic transition period from an old industrial region to new sectors.

In prior studies related to residents’ support for tourism, emotion, and sense of place were the few affective attitudes that could be directly influenced by perceived tourism impacts [30,75]. However, we have provided explicit evidence that perceived tourism impacts can directly influence place attachment. In the current study, the relationship was examined and the results indicated that if the development of IHT induces positive impacts that are perceived by residents, the residents will develop stronger functional and emotional attachment to the place. This can be explained by some indirect evidence that the more positive the residents consider the environment around them, the more functional the bonds (place dependence) attached to the environment [10], and if residents enjoy the living conditions and social cultural environment of their city, they will develop more attachment to it [9].

The construct “perceived negative tourism impacts” was only found to be related to place dependence; an influence on place identity was not found. This indicates that the perception of negative impacts from the development of IHT will only weaken the level of residents’ functional attachment to a place and will not change their emotional bonds to a place. A possible explanation for this is related to the transition to IHT. As a new direction for local development, the tourism industry is in its infancy. It has more influence on the change in urban function (such as the establishment of new scenic spots and the reuse and redesign of industrial heritage). Therefore, the negative impacts of tourism development can play crucial roles in residents’ functional bonds to the city. As resident’s emotions toward their living place are relatively stable [1,61], the assessment of negative tourism impacts may only have a limited influence on residents’ emotional attachment to the place, especially when the positive assessment of tourism development outweighs the perceived negative impacts (Appendix A).

Although we tried to examine whether place identity and place dependence (not a unidimensional construct of place attachment as in Stylidis [10]) could both influence residents’ support for tourism development, the results only showed that place identity significantly and positively influences residents’ support for tourism, indicating that the more Huangshi means to local residents, the more positive their support IHT development, consistent with studies by Wang and Xu [9] and Schuster et al. [72]. However, place dependence showed no significant influence on residents’ support for tourism. This can be explained by place identity being a more direct predictor than place dependence in interpreting community sustainability in the context of tourism [72]. Residents’ deep memories about Huangshi are consistent with the city’s cultural and historical connotations. IHT with the same content can arouse residents’ resonance, leading to a close relationship between residents’ emotional bonds and their support for tourism.

Prior studies have proven that the multidimensional construct of tourism involvement can influence the perceived tourism impacts [82,83] and place attachment [24,85]. In this study, the findings confirmed that a unidimensional construct of involvement could also be a determinant predictor of cognitive (perceived impacts) and affective (place attachment) attitudes. The more motivation, arousal, and interest residents have in relation to IHT, the more positive their assessments of it. With the deepening level of involvement in tourism, more details about the negative impacts of tourism will be understood by residents. These results are in line with the findings of Lee [47], Choi and Murray [82], and Sebele [84]. Contrary to prior findings [47], residents’ involvement in IHT was not shown to significantly affect their support. A possible explanation for this is that residents usually do not become involved in management processes or the planning of tourism development [83]. Additionally, tourism involvement was shown to positively influence place identity and dependence. If residents become more involved in IHT, their emotional and functional bonds to Huangshi would deepen. The results fill a research gap by showing that a unidimensional construct of involvement can affect a multidimensional construct of place identity and place dependence.

Based on the above findings, some policy recommendations are provided to improve resident support for IHT industry in resource-exhausted cities. The first recommendation is to improve residents’ place identity in their city. Because the findings show that residents’ place identity is also a crucial determinant of their support for the transition and development of hosting tourism in their community. This can be achieved by establishing emotional bonds between residents and the place through more interactions around the theme. For example, in relation to Huangshi, planners should use the nostalgic attributes and contents (such as industrial history and traditions in the past) of industrial heritage sites to stimulate resident’s memory of and emotion toward Huangshi [37,75]. Local authorities can highlight the contribution of residents to the development of the city’s physical environment through media, and provide appropriate incentives to enhance the relationship between residents and the city with regard to maintain residents’ long-term support. Among the respondents in this study, about 22.1% had lived in Huangshi for less than 10 years, thus, local authorities should also encourage them to live there longer, because the length of residence can also influence place attachment [63].

Another key point is to make residents feel like they can obtain more significant benefits from tourism development by maintaining and maximizing the positive impacts, while decreasing and controlling the negative impacts within a reasonable range. Considering the studies related to residents’ support for tourism development [1,2,30,44,46], the following policy recommendations are proposed. Firstly, as most destinations face similar negative impacts, such as increased living costs, crime, pollution, and environmental damage, destination authorities and developers should actively manage these problems. Otherwise, they will lose local support, blocking sustainable tourism development [30]. Secondly, to promote the perceived positive impacts, the key is to make residents receive concrete benefits. Destination managers could highlight some iconic contributions around residents’ working and living environments, such as convenient transportation, more job opportunities, and good air quality. Thirdly, the above recommendations can only be used as a short-term measure. Enhancing residents’ trust in the government and empowering them are the fundamental methods to manage the perceived impacts and gain more support for sustainable tourism development in the long run [2,20].

For cities such as Huangshi that are economically transitioning from an old industrial region to a tourism destination in China, the sustainability of the tourism industry determines the success of the economic transition. The findings show that involvement can shape perceived impacts and place attachment. Therefore, the final policy recommendation is to strengthen residents’ involvement in local tourism. Local authorities and sites managers are encouraged to develop strategies that attract more residents to participate in IHT to help them form perceptions of positive impacts, deepen place attachment, and build a sustainable tourism industry. Improving the values of industrial heritage can be a driver of community participation in heritage tourism [116]. For example, in the case of Huangshi, competitions such as photography and landscape modeling design for industrial heritage enthusiasts can be held regularly. For local students, lectures and academic conferences about the industrial history of sites are options. For citizens who are interested in leisure activities, an industrial design facility can also be the right choice to attract their attention to industrial heritage sites.

As countries in the world have different conditions (e.g., level of education, national wealth, democracy, investment policy), for other industrial cities in developing countries, the results can only give some macro recommendations for their transitions. The transition to IHT is an effective method for developing continuous industries and emerging substitute industries, since the difficulty and cost of implementing the tourism transition are relatively small. In the initial stage of the transition, the advantages of local resources (such as industrial facilities and culture) can be brought into full play and the implementation cycle is relatively short. Furthermore, the transition to IHT can be seen as a new practice of resource and environmental protection, which will be beneficial to the regeneration and establishment of the new image of the industrial cities. It can be proven by the most concerned perceived positive impacts (improving Huangshi’s holistic image and environmental conservation and protectionism) as shown in Appendix A. Above all else, when residents feel the benefits from the tourism transition, they will support IHT development. This will determine the sustainability of the tourism transition progress.

There are several limitations to this study. Firstly, direct effects of tourism involvement on perceived tourism impacts and place attachment exist, and perceived tourism impacts and place attachment also directly influence residents support for tourism. Tourism involvement does not directly influence residents’ support for tourism. Therefore, according to the mediation definition [117], indirect effects of perceived tourism impacts and place attachment between tourism involvement and residents support for tourism might exist. Future studies could examine these indirect effects to stimulate further discussion. Secondly, the operationalized constructs of place attachment and involvement are still not sufficiently comprehensive. Place attachment was operationalized into two constructs (place identity and place dependence) [24], and involvement was operationalized into a unidimensional construct [21]. As the two constructs of place attachment may overlook the interactive effects between a place’s physical contents and its residents’ personalized emotions [33], more dimensions, such as social bonding and affective attachment, could be added to fill in the research gap in future studies [67]. The assessment of residents’ holistic involvement in tourism may ignore its multifaceted nature suggested by prior studies (such as pleasure, risk, and consequence) [77,78]. Therefore, further testing of the multidimensional constructs of the model proposed in this study is needed to explore the multiple impacts of involvement. Thirdly, the sampling period was inside the tourism season of the Spring Festival. People usually have a pleasant mood and are passionate during the annual national event. Lawler and Thye [118] suggested that (non-rational) passion could influence people’s attitudes and behavioral intentions in the context of SET. This unique factor is like an invisible motivator that gives residents more energy and passion and may indirectly influence the relationships between the constructs in this study. Therefore, in future studies, methodologies such as variance analysis, multiple-group analysis, and moderating analysis are potential choices to determine how and why the unique factors (e.g., the passion originating from the festival atmosphere) affect the residents’ attitudes toward tourism development. Fourthly, although the number of respondents (N = 336) matches the basic statistical requirements for sample size, it may not ensure that enough residents had been included in the sample, since Huangshi has a population of 2.68 million (as of 2017) [86]. Power analysis could be a good choice for follow-up studies to determine a more feasible number of respondents [119]. Finally, since Huangshi is still in the initial stage of IHT development and the perceived impacts may change over various stages, it would be hard to measure the real tourism impacts in the short run. More precisely, it might be better to rephrase it into potential or expected impacts for the respondents who live in the city. To further examine the effects of the model proposed in this study, a long-term follow-up survey (e.g., every five years) could be a suitable option.