How Visitors Perceive Heritage Value—A Quantitative Study on Visitors’ Perceived Value and Satisfaction of Architectural Heritage through SEM

: In a time when heritage conservation is coordinated with tourism and urban renewal to achieve sustainable development, value is considered as the intrinsic factor of heritage protection, utilization, and management. From the perspective of visitors, this article aims to delve into the key factors of the perceived value of architectural heritage and their relationship with the value types, via structural equation modeling (SEM). This article constructs a research model of the perceived value, perceived enjoyment, visit satisfaction, and post-visit behavioral intention of architectural heritage, based on the American Customer Satisfaction Index (ACSI). The data for the analysis were 271 eligible questionnaires obtained from an anonymous random sample of visitors to architectural heritage sites in Guangzhou, Guangdong (Canton) Province, China. The results show that visitors’ perception of architectural heritage value does not follow the value types proposed by experts, but is based on a comprehensive assessment of personal perceptions. Meanwhile, visitors’ perceived value mainly depends on the artistic value, although the historical value is recognized by experts as dominant, and visitors prefer historical stories that are intertwined with culture. The results also reveal that, as with the perceived value, the perceived enjoyment, which was rarely considered in heritage conservation and management in the past, has a considerable positive impact on visit satisfaction, and can also improve visitors’ perceived value. The proposed theory model and research results can serve as a valuable reference for cultural heritage management and for operators of cultural heritage tourism destinations, and can also provide new ideas and methods for heritage value research.


Introduction
With the increasing level of urbanization, China has fully entered the stock era, and the construction direction has changed to optimize urban resources, enhance urban quality and improve cultural life. Various built heritage sites, especially architectural heritage sites, have become well-received tourist attractions and are among the core drivers of urban stock renewal and development [1].
Accordingly, in September 2021, the Opinions on Strengthening the Protection and Inheritance of History and Culture in Urban-Rural Construction (hereinafter referred to as the Opinions) was issued, emphasizing the rational utilization of the architectural heritage sites, integrating protection and inheritance into current economic and social development, ecological civilization construction, and modern life [2]. It is generally accepted that heritage conservation can be compatible with heritage consumption, and sustainable cultural heritage tourism is one of the effective strategies to balance conservation, financial support, and public access [3].
Architectural heritage not only attracts tourism and drives tourist-derived revenue, but also brings about broader economic, social, and environmental benefits. However, the future of tourism is likely to see significant development through innovative management 1.
How do visitors experience the value of an architectural heritage site? 2.
Is their perception of value affected by the value types proposed by experts? 3.
What type of value is most attractive to visitors? 4.
What is the relationship between the perceived value and the perceived enjoyment? 5.
Are visit satisfaction and the post-visit behavioral intention determined by the perceived value and the perceived enjoyment?

Visit Satisfaction
Generally speaking, satisfaction refers to the subjective evaluation of a product or service in terms of meeting customer needs and expectations [12], or the overall level of satisfaction of tourists with the tourism object and process [13]. The chain of perceived value-satisfaction-loyalty (or post-visit behavioral intention) has been repeatedly verified by a large number of studies with reference to four major international theories: the American Customer Satisfaction Index (ACSI) [14], the European Customer Satisfaction Index (ECSI) [15], the evaluation model and the Service Quality Model (SERVQUAL) [16], and the measurement dimensions of the China Tourism Institute on Tourism Satisfaction Survey (China's domestic research) [17].
Based on the ACSI, this article constructs a research model of the perceived value, visit satisfaction and post-visit behavioral intention regarding architectural heritage sites. Perceived value has an indirect effect on behavioral intention through the intermediary of satisfaction. Satisfaction is the result of perceived value, while post-visit behavioral intention is the final outcome variable [18].

Perceived Value
The perceived value is a combination of economic and time cost, and physical and spiritual benefit [19], which has been proved to be the crucial element in cultural heritage planning and heritage tourism management, as it strongly influences tourists' satisfaction [20].
Research on perceived value can usually be separated into various variables corresponding to the characteristics of products and services [21]. For example, it can be considered as all or a few of the following variables: efficiency, excellence, status, esteem, play, aesthetics, ethics, and spirituality [22]. It can also be separated into two variables: utilitarian value and hedonic value [23]. The Global Visitor Perceived Value Scale includes six variables: facility functional value, professional skills of travel agency personnel, package tourism product quality, functional value, emotional value, and social value [24]. In terms of built heritage, Chinese scholars have also designed various "perceived value" dimensions (see Table 1).
According to the value characteristics of architectural heritage, this research aims to explore the "function" and "entertainment" variables in the "perceived value", as well as the influencing mechanism on satisfaction and post-visit behavioral intention. In order to distinguish it from the following architectural heritage value, this study named the two variables of "function" and "entertainment" as "perceived value" and "perceived enjoyment", respectively.  [29] Emotional value, tourism resources perception, cognitive value, social value, perception of non-economic costs, perception of tour guide service, perception of economic cost, perception of community service

Value Types of Architectural Heritage
Over the past few decades, heritage value has become one of the paramount issues in discussions on conservation and restoration, within a multicultural context [30]. Mason has proposed a vivid analogy, that is, these value typologies tend to describe the same pie sliced in different ways to depict the diversity of heritage value typologies constructed by various scholars and organizations over time and across different cultures [31]. This article defines the perceived value as the set of perceptual value of all the attributes encountered throughout the visit [32].
On the one hand, unlike most existing tourism-based research, this article aims to establish a more direct connection between the value characteristics of architectural heritage and the variable of "perceived value", and to reveal visitors' perceptions of popular heritage value types and their contributions to visitors' overall perceived value through quantitative analysis. Scholars have many definitions of the same or different types of value, among which the most popular ones are recognized artistic (aesthetic) value, historical value, cultural value, and scientific (technical) value [33].
On the other hand, these four types of value are the main elements of visitors' consumption and experience. Meanwhile, the value types should be distinguished clearly and be simple enough to avoid obscuring the connotations of general visitor questionnaires. For example, we chose cultural value and excluded social value because there is too much overlap between them, and also because economic value is too difficult for laypersons to evaluate, while emotional value can easily lead to different understandings.

Materials and Methodology
There are three steps in this study. The first step was to construct a theoretical framework (structural equation model) based on the fully validated ACSI model and to propose hypotheses. The second step was to process the qualified questionnaire and analyze the data through SPSS 19 and AMOS 23. The third step was to verify the theoretical model through AMOS 23 and analyze the key influencing factors.

Theoretical Framework
The structural equation model (SEM) is a statistical research method, which is widely used to analyze linear relationships between observed and latent variables, or between latent variables. It has been widely applied in tourism research to investigate constructs such as value, satisfaction, and post-trip behavior intention, yielding substantial empirical support [5,6,18]. In terms of product and service satisfaction, a research model is built by extending the American Customer Satisfaction Index (ACSI) according to the characteristics of heritage tourism [23]. The ACSI is flexible enough to be applied to different research objectives, so it is widely used in extensive fields, including heritage tourism and management [34]. The research model of this study is as follows (see Figure 1). heritage and the variable of "perceived value", and to reveal visitors' perceptions of popular heritage value types and their contributions to visitors' overall perceived value through quantitative analysis. Scholars have many definitions of the same or different types of value, among which the most popular ones are recognized artistic (aesthetic) value, historical value, cultural value, and scientific (technical) value [33]. On the other hand, these four types of value are the main elements of visitors' consumption and experience. Meanwhile, the value types should be distinguished clearly and be simple enough to avoid obscuring the connotations of general visitor questionnaires. For example, we chose cultural value and excluded social value because there is too much overlap between them, and also because economic value is too difficult for laypersons to evaluate, while emotional value can easily lead to different understandings.

Materials and Methodology
There are three steps in this study. The first step was to construct a theoretical framework (structural equation model) based on the fully validated ACSI model and to propose hypotheses. The second step was to process the qualified questionnaire and analyze the data through SPSS 19 and AMOS 23. The third step was to verify the theoretical model through AMOS 23 and analyze the key influencing factors.

Theoretical Framework
The structural equation model (SEM) is a statistical research method, which is widely used to analyze linear relationships between observed and latent variables, or between latent variables. It has been widely applied in tourism research to investigate constructs such as value, satisfaction, and post-trip behavior intention, yielding substantial empirical support [5,6,18]. In terms of product and service satisfaction, a research model is built by extending the American Customer Satisfaction Index (ACSI) according to the characteristics of heritage tourism [23]. The ACSI is flexible enough to be applied to different research objectives, so it is widely used in extensive fields, including heritage tourism and management [34]. The research model of this study is as follows (see Figure 1).

Hypotheses
Based on the literature review and interviews with five experts on architectural heritage conservation and management from South China University of Technology, hypotheses are proposed in this paper. Figure 1 shows all assumed path coefficients and their correlations. The details of these hypotheses are explained as follows.

Questionnaire Design
For built heritage, questionnaire surveys and statistical analysis are recognized as the most effective quantitative methods to study the perception and experience of a large number of individuals, and also to more reliably verify the proposed hypotheses [35]. Accordingly, a questionnaire was designed to identify latent variables that contribute to the overall satisfaction with the heritage visit.
The questionnaire consists of two parts: basic demographic information and latent variables. The demographic part includes names and order scales with 5 items: gender (BD1), age (BD2), education level (BD3), residential area (BD4), and monthly household income (BD5). The latent variables contain (1) artistic value; (2) historical value; (3) cultural value; (4) scientific value; (5) perceived value; (6) perceived enjoyment; (7) visit satisfaction; (8) post-visit behavioral intention. Most of the questions on each variable are adapted from well-validated scales, and the expression of the questions has been appropriately adjusted according to the research content (see Table 2). For example, the questions about the perceived value are adapted from studies by Wang [36] and Boo, Busser and Baloglu [37]. For this part, respondents use a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree) to express their agreement or disagreement.
To confirm the standardization and validity of the questionnaire and each item, we first consulted 5 related experts in the field of architectural heritage from South China University of Technology, and then conducted a pilot survey of 50 visitors in the Guangzhou Guangxiao Temple. Based on the analysis of the reliability and validity of the pre-survey, the questionnaire was revised until all measurement items performed well and reached acceptable levels (Cronbach's Alpha > 0.8, KMO > 0.75) [38].

Data Collection
To test the hypotheses, Guangzhou (Guangdong/Canton province, China) was selected as the research area. Guangzhou was identified as one of the first batch of national historical and cultural cities, with a rich architectural heritage. At the same time, Guangzhou is one of the most economically developed cities in China, and ranks among the top in China, in terms of heritage management, revitalization, and tourism (see Figure 2). This study selected four architectural heritage sites for investigation: Guangzhou Guangxiao Temple, Chen Clan Ancestral Hall, Sacred Heart Cathedral, and Shamian Architectural Complex, all of which are located in the old town of Guangzhou and have been identified as national cultural relic protection units in China. They respectively correspond to the mainstream architectural heritage site types in China: the traditional temple, traditional ancestral hall, colonial architecture, and architectural complex.
The survey took place on 1 October, 6 October, 17 October, and 26 October 2019, at  This study selected four architectural heritage sites for investigation: Guangzhou Guangxiao Temple, Chen Clan Ancestral Hall, Sacred Heart Cathedral, and Shamian Architectural Complex, all of which are located in the old town of Guangzhou and have been identified as national cultural relic protection units in China. They respectively correspond to the mainstream architectural heritage site types in China: the traditional temple, traditional ancestral hall, colonial architecture, and architectural complex.
The survey took place on 1 October, 6 October, 17 October, and 26 October 2019, at the exit of the sites. The investigators were trained second-and third-grade undergraduates majoring in architecture. The data collection method was simple random sampling, face-toface interviews, and on-site completion of surveys. Respondents were visitors aged 12 and above, and the number of them was determined by the size and internal homogeneity of visitors when they arrived.
The ratio of observed variables (all latent variable items) to the number of questionnaires in the model is generally 1:10-1:15 [47]. Since there are 24 observed variables in the research model, the valid questionnaire sample must be more than 240. In this study, 360 questionnaires were distributed and 360 were returned, of which 271 were valid and the rest were excluded. The quantity of questionnaires met the requirements of SEM analysis. The demographic characteristics of the visitor sample are shown in Table 3.
It is worth noting that (1) the survey was conducted in 2019, prior to the outbreak of the COVID-19 pandemic; (2) due to the significant differences between architectural heritage tourism sites and general tourism sites, such as the former being more informative than entertaining, there are distinct demographic characteristics of surveyed groups that differ from those of ordinary tourists. For instance, the visitors surveyed tended to be younger, with 57.6 percent aged between 15 and 25. This is partly due to the higher non-response rate among older visitors, while younger individuals were more likely to participate in the survey; (3) out of 271 questionnaires, only two were completed by foreign visitors. Despite being an international city with a large number of foreign visitors, Guangzhou also has a particularly high ratio of Chinese to foreign visitors. According to the Guangzhou Yearbook (2015), there were 3,002,600 foreign tourists in 2014 out of a total of 162 million tourists, accounting for only 1.9% [48].

Analysis and Results
The SEM analysis consisted of three steps: (1) the reliability and validity verification; (2) the fitness assessment of the research model; (3) the analysis of key factors and path of the structural model [49].

Reliability and Validity Analysis
The SEM studies should validate the variables and items of the research model before conducting the analysis. Confirmatory factor analysis (CFA) is a recognized method to verify reliability and validity. Therefore, this study started with CFA analysis of all variables and items.
The eight variables of the SEM are shown in Table 4. The factor loading (Std.) of all items is between 0.6 and 0.9, except for the AV3 item of the artistic value variable, and the factor loading of AV3 is also above 0.5. Therefore, the factor loadings of each item in this research model met the standards. Secondly, the composition reliability (CR) of each item is between 0.7 and 0.9, indicating that they had sufficient internal consistency. Thirdly, the convergence validity (AVE) of each item is greater than 0.5, which proves that the convergence validity among the items met the requirements [14,47].

Fitness Analysis of Research Model
The fitness assessment of the research model is an inevitable process of SEM analysis. The higher the fit degree is, the closer the model matrix is to the sample matrix. In reference to the research results of the studies of McDonald and Ho [50], and Jackson, Gilasyp and Stephenson [51], several key indicators were used to evaluate the overall fitness of the research model, including the chi-square (χ 2 ), the ratio of chi-square/df (degree of freedom), the comparative fit index (CFI), the goodness-of-fit (GFI), the adjusted fit index (AGFI), and the root-mean-square error of approximation (RMSEA).
As shown in Table 5, the ratio of chi-square degrees (χ 2 ) of freedom (df) is 2.778, which is less than 3, indicating that the fit of the model to the sample data was adequate. The goodness-of-fit (GFI) is 0.838 and the comparative fit index (CFI) is 0.890, both greater than 0.8. The adjusted goodness-of-fit index (AGFI) is 0.795, slightly less than 0.8. The RMSEA demonstrates a difference of 0.081 between the variance-covariance matrix acquired from the sample data and the variance-covariance matrix composed of the theoretical model [38], which is slightly higher than the ideal index of 0.08.
Nevertheless, Bentler and Chou noted that it is arduous for all indicators to fully meet the generally good fitness with a multivariate model [52]. This paper further deconstructed the perceived value variables into the artistic value, historical value, cultural value, and scientific value of the architectural heritage, so as to connect visitors' perceived value more deeply and directly with the architectural heritage value types, and improve the practical significance of the research findings. However, the fitness index of a second-order model must be lower than that of the first-order model [51], so an RMSEA of 0.081 is acceptable. Therefore, the overall fitness of the research model is acceptable.
This study verifies the hypothesized variable relationships by examining the path coefficients between the latent variables and combining the p value to determine whether the hypotheses hold. The model standardization path is shown in Figure 3, and the hypothesis test results are displayed in Table 6.   As can be seen from Table 6, the parameter estimates of H1, H2, H3, H5, and H6 in the research model reached the significance level of p < 0.01, and the parameter estimate of H8 is 0.012, also having reached a significant level of p < 0.05, indicating that these hypotheses were supported by the sample data. The parameter estimates of H4, H7, and H9 did not reach the significant level of p < 0.05, which proves that these hypotheses were not supported by the sample data.

Key Factor and Path Analysis
As shown in Figure 4, the path analysis results among variables in the research model indicate the key factors influencing visitors' perceived value, satisfaction, and post-visit behavioral intention of the architectural heritage:
Perceived value (β = 0.510|p < 0.01) is the second most important factor that influences the visit satisfaction, slightly lower than the perceived enjoyment; 3.
Visit satisfaction (β = 0.681|p < 0.01) has a significant impact on post-visit behavioral intention, including revisit and word-of-mouth communication; 4.
The standardized path coefficient of perceived enjoyment on behavioral intention is 0.138, p = 0.072 > 0.05, indicating that the perceived enjoyment has no obvious influence on post-visit behavioral intention, and there is no direct relationship between the two; 5.
Perceived enjoyment (β = 0.231|p < 0.01) is the third most important factor that effects the perceived value; 8.
The standardized path coefficient of historical value to perceived value is 0.057, p = 0.615 > 0.05, indicating that the historical value of architectural heritage has no significant influence on the perceived value of visitors; 9. The

Correlation Test among Variables of Heritage Value Types
It is undeniable that the definitions and boundaries of the various value types of heritage are ambiguous and overlapping to some extent, which is more likely to cause different understandings of non-experts [31]. However, it is also an important task for this paper to explore the cognition of non-experts on the value types of heritage.
As shown in Table 7, the correlation tests among the variables of heritage value types are as follows: 1. There is a significant correlation between the four variables of architectural heritage value types (artistic, historical, cultural, and scientific value); 2. The standardized correlation coefficient between artistic value and scientific value is the highest (0.663), followed by that with historical value (0.576); 3. The standardized correlation coefficient between historical value and cultural value is the highest (0.77), followed by that with scientific value (0.536); 4. The standardized correlation coefficient between cultural value and historical value is the highest (0.77), followed by that with scientific value (0.525).

Correlation Test among Variables of Heritage Value Types
It is undeniable that the definitions and boundaries of the various value types of heritage are ambiguous and overlapping to some extent, which is more likely to cause different understandings of non-experts [31]. However, it is also an important task for this paper to explore the cognition of non-experts on the value types of heritage.
As shown in Table 7, the correlation tests among the variables of heritage value types are as follows: 1.
There is a significant correlation between the four variables of architectural heritage value types (artistic, historical, cultural, and scientific value); 2.
The standardized correlation coefficient between artistic value and scientific value is the highest (0.663), followed by that with historical value (0.576); 3.
The standardized correlation coefficient between historical value and cultural value is the highest (0.77), followed by that with scientific value (0.536); 4.
The standardized correlation coefficient between cultural value and historical value is the highest (0.77), followed by that with scientific value (0.525).

Analysis Based on Demographic Characteristics
Since the validity of the combination of items and variables in the research model has been verified, Pearson correlation analysis was performed for demographic items and the first item of each variable through SPSS 19.0. The results are shown in Table 8. 1.
BD1 has a correlation with SV1 (0.135|p < 0.05), indicating that females perceive the scientific value of architectural heritage sites as better than males; 2.
BD3 has a correlation with PE1 (−0.138|p < 0.05), demonstrating that visitors' education level has a significant negative correlation with their perceived enjoyment. Moreover, BD3 has a correlation with BI1 (−0.183|p < 0.01), explaining that visitors' education level also has a significant negative impact to the post-visit behavioral intention; 4.
BD5 has a correlation with HV1 (−0.137|p < 0.05), demonstrating a significant negative correlation between visitors' monthly household income and their perception of the historical value; 5.
BD4 has no correlation with the variables in the research model. Note: ** Significantly correlated at the 0.01 level (bilateral). * Significantly correlated at the 0.05 level (bilateral).

Perceived Value and Value Types
For one thing, the Burra Charter advocates a value-oriented dynamic process of change management for cultural heritage, and the China's Opinions Charter also explicitly takes the value-oriented as the most basic working principle of urban and rural heritage conservation. In a quantitative way, this study again verifies that perceived value is a key factor that significantly affects visit satisfaction and post-visit intention.
For another, in the sphere of built heritage conservation, experts usually use value types for value evaluations, which can be traced back to the studies of the ICOMOS Bara Charter of Australia [53], and it has become the most popular method in the 21st century [54]. However, a growing number of scholars are reflecting and questioning the value type groups and the Authorized Heritage Discourse led by experts. They point out that, in marked contrast to experts, non-experts do not perceive heritage value through value types "designed" by experts, but rather, tend to perceive heritage value through emotional experiences [55]. The statistical analysis results in this article found similar phenomena: 1.
The significant correlation between the artistic, historical, cultural, scientific values indicates that most visitors are not sensitive to the types of architectural heritage value types set by experts. In other words, they do not perceive the value of architectural heritage sites according to the value types, but tend to assess them with their overall feelings;

2.
The artistic value is the first factor that has a significant positive impact on the perceived value, revealing that visitors' perception of the architectural heritage value of a site mainly depends on its artistic value, that is, the artistry, excellence, and innovation of architectural design and construction; 3.
The standardized correlation coefficient between the artistic value and the scientific value is 0.663, which means that for visitors, the perceived scientific value still belongs to artistic value to a large extent; 4.
Cultural value is the second significant factor influencing the perceived value, with a standard path coefficient of 0.280, while the standard path coefficient of historical value to perceived value is only 0.057. Meanwhile, the standard correlation coefficient between cultural value and historical value is 0.770.
These results illustrate that although historical value has been recognized by experts as the most important value of architectural heritage sites from the very beginning, nonexperts are not sensitive to the "pure" historical value, and they prefer historical information intertwined with culture, such as historical stories.

Perceived Enjoyment
In plenty of expert-led architectural heritage conservation projects, the physical situation of architectural heritage sites is usually the core, while visitors' perceived enjoyment in the heritage experience is often neglected or underestimated. However, the research results prove that both the perceived enjoyment and perceived value have a significant positive impact on visit satisfaction, and the standardized path coefficients of perceived enjoyment is even slightly higher, which are 0.536 and 0.510, respectively. This demonstrates that the entertainment experience is as important as the perception of heritage value, both of which are indispensable to visit satisfaction.
More importantly, the perceived enjoyment also has a significant positive effect on the perceived value, with a path coefficient of 0.231, which indicates that entertainment experiences can promote deeper perceptions of heritage value. Therefore, architectural heritage conservation planning and management should not only focus on the restoration of the physical elements, but also fully consider the psychological needs of people in current society and communities, and their entertainment experience of the heritage visit.

Analysis Results Based on Demographic Characteristics
Two noteworthy phenomena are derived from the analysis results based on the demographic characteristics of the questionnaires:

1.
Age has a significant positive correlation with the historical value, cultural value, perceived enjoyment, visit satisfaction, and post-visit behavioral intention. The standardized correlation coefficients are 0.143, 0.153, 0.167, 0.186, and 0.277, respectively. However, most of the visitors are young, with 83.1% of them aged 15 to 34. This indicates that in the current architectural heritage projects, there is a lack of lively, cuttingedge, and interesting displays and facilities for display that appeal to young people. As young people have become the main visitors to the architectural heritage sites, there is an urgent need for both conservation practitioners and government officials to be more proactive in considering their market and making targeted adjustments.

2.
The visitors' level of education has a significant negative correlation to the perceived enjoyment and post-visit behavioral intention. The standardized correlation coefficients are −0.138 and −0.183, respectively. Among the visitors, the proportion with an undergraduate degree is the highest, accounting for 53.5%, followed by college degree (20.3%), and 81.2% of the respondents have a college degree or above. This shows that the current display and management methods of architectural heritage sites still need to be improved-targeting young people with a higher level of education, adopting more innovative and interactive ways of display, and enhancing the entertaining experience of the visiting process and the display content should focus more on the artistic and cultural value.

Conclusions
Multi-party participation in heritage protection and management has become a basic working principle at home and abroad. Heritage conservation experts, managers, and practitioners should provide more access to professional knowledge to laypersons, and grasp how laypersons actually perceive heritage values and how they are satisfied with their heritage visit.
From the perspective of visitors, this article constructs a research model of the perceived value, perceived enjoyment, visit satisfaction, and post-visit behavioral intention regarding architectural heritage sites, based on the American Customer Satisfaction Index (ACSI). Through the analysis of 271 eligible questionnaires, this study has revealed fascinating insights into the behavior of visitors and their perceptions of heritage sites.
Artistic and cultural values are of greater significance than scientific and historical ones. The latter, however, are commonly cited as the primary reason for designating such sites as heritage sites worldwide. This implies that visitors to heritage sites are more interested in being "impressed" and "shocked" than "educated". Therefore, it is crucial to explore the artistic and cultural value of cultural heritage sites, as well as to build strong emotional connections between contemporary visitors and heritage sites in order to effectively protect and utilize them.
The findings also suggest that the perceived enjoyment, which was previously overlooked in heritage conservation and management, has a significant positive impact on visitor satisfaction and even the perceived value.
The results presented in this paper can serve as a valuable reference for cultural heritage management and for operators of cultural heritage tourism destinations to creatively and positively respond to the feelings and expectations of visitors and facilitate the realization of heritage tourism benefits and sustainable development.

Limitations and Future Research
Despite the theoretical and practical contributions, there are some limitations in this paper. Firstly, in terms of sample group characteristics, 83.1% of visitors were 34 years old or younger. On the one hand, this is because of the uniqueness of architectural heritage tourism displays, such as low entertainment and high information. On the other hand, this is also because older people are more likely to reject the questionnaire. For example, some older people declined the questionnaire because they could not see clearly or had trouble reading. Some middle-aged people rejected the survey, saying they were in a hurry. Young people, by contrast, tended to be more interested in questionnaires and academic research and were more willing to participate. In the future, we will also need to diversify research methods to make them more age-friendly to include older people into study groups.
Secondly, this article deconstructs the perceived value variables into four secondorder variables of architectural heritage, thus simplifying the research model. However, simplifying the behavior of architectural heritage visiting also has some shortcomings; for example, the model fitting index will be relatively low.
Finally, the four cases selected in this article cannot fully cover all types of architectural heritage, and meanwhile, the four architectural heritage sites are all in Guangzhou, China. With continuous advancements of economy and urban renewal, architectural heritage sites will not only be regarded as popular tourist attractions, but also important places for leisure, entertainment, and cultural life in cities. Therefore, the expectation, behavior, and satisfaction mechanisms of architecture, urban, landscape, and other built heritage sites still need to be explored in a more in-depth and comprehensive way in China and other cultures and areas. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.