Product or brand familiarity has frequently been linked to several past product-related experiences that consumers have accumulated (Stylidis, 2020), and it is one of the most important variables in the marketing literature (Kim et al., 2019). Within the literature on tourism, destination familiarity has always been treated as a single-dimension construct measured through previous destination experience (Hu and Ritchie, 1993), which can be used to contrast first visits and revisits (Prentice, 2006). In other words, destination familiarity is the basis for explaining differences in various aspects of travel behavior between the images of a given tourist destination of visitors and non-visitors (Andreu et al., 2000). Researchers have reported that many visitors feel secure in a familiar destination, while novel destinations are likely to seem riskier to visitors (Lepp and Gibson, 2003). Therefore, destination familiarity allows researchers to understand how visitors shape their mental image of a particular destination (Chen and Lin, 2012).

Nevertheless, researchers have criticized the measurement of the single-dimension construct of destination familiarity, noting that is not appropriate in the temporal tourism environment (Stylidis, 2020; Ju et al., 2021). According to Kim et al. (2019), destination familiarity can be increased by direct experiences, such as purchasing and usage behavior, while indirect experience can also significantly influence the familiarity of a tourist with a given destination; this can include exposure to advertisements, reviews from others, the geographic distance between origin and destination, having received information regarding a given destination, perceived knowledge related to education, media coverage, travel guides, social media, and personal contact with others (Kim et al., 2019; Chi et al., 2020; Stylidis, 2020; Ju et al., 2021). Researchers have therefore suggested that two aspects of destination familiarity should be operationalized: experiential and informational (Kim et al., 2019). Experiential destination familiarity refers to the level of awareness created through past travel experience, whereas informational familiarity is accumulated by indirect exposure to a destination (Baloglu, 2001). The results of many studies have demonstrated that destination familiarity based on past travel experience significantly affects the travel decision-making process of visitors (Prentice, 2006; Tan and Wu, 2016), but visitors can still attain a definite level of destination familiarity via indirect experience (Chi et al., 2020). Other means of direct experience in creating destination familiarity have largely been overlooked in tourism literature (Ju et al., 2021).

The formation of DVI is among the factors that play a significant role in creating tourist loyalty (Králiková et al., 2020). The importance of this concept has led to a growing body of studies on tourism destinations (Alcocer and Ruiz, 2019), and the findings of these studies indicate that DVI has a significant impact on loyalty, satisfaction (Ragab et al., 2019), visit intention (Wang et al., 2021a), and revisit intention (Melo et al., 2017). However, there is currently no agreement on a precise definition for this concept (Ragab et al., 2019) because previous studies have defined DVI in many different ways. Prior studies defined DVI by incorporating a set of feelings, emotions, and attitudes that tourists hold regarding the destination (Ragab et al., 2019). The most dominant definition of DVI in previous studies is the overall impression of tourists on a specific destination based on their knowledge and feelings about that destination (Ragab et al., 2019). However, DVI should be considered a multidimensional concept with three primary dimensions: cognitive, affective, and conative (Králiková et al., 2020). The beliefs and knowledge about a particular destination belong to the cognitive part; interactions with and feelings toward a particular destination belong to the affective image; the conative dimension evolved from the cognitive and affective images and is considered to be analogous to behavior. Both cognitive and affective images are the best predictors of destination visitation (Králiková et al., 2020).

The influence of tourist perception, DVI, and satisfaction has been a topic of great research interest in tourism (Rajesh, 2013). Over time, DVI piqued the curiosity of both practitioners and academics, and it has been examined in relation to the destination brand since the 1990s (Almeyda-Ibáñez and George, 2017). The image can be described as “the feelings of people of anything that they were aware” (Boulding, 1956, p. 15) and “a way of organizing the different stimuli received on a daily basis and help make sense of the world in which we live in” (Mayo, 1973). Numerous researchers have asserted that DVI is only one component of the destination brand (Konecnik and Gartner, 2007, cited in Herle, 2018). Overall, DVI can be characterized in two ways: the affective image, in which the destination image is formed based on the beliefs and feelings of tourists toward the destination; and the cognitive image, which is created as a result of the assessment of the tourists of the historical, social, political, and economic characteristics of the destinations (Wang et al., 2021a). Destination visit image is critical to the decision-making process of tourists because a destination that lacks a favorable or positive image will find it difficult to compete (Králiková et al., 2020). The images of tourists of a destination can be quite personal and subjective because tourists can form their views based on individual perceptions of the place (Alcocer and Ruiz, 2019). Therefore, DVI is not only important regarding tourist satisfaction within a given destination but also influences the behavior of tourists, including on-site experiences, evaluations, and destination loyalty (Králiková et al., 2020). Alcocer and Ruiz (2019) reported that DVI highly influenced the satisfaction of tourists toward a cultural destination, while Ragab et al. (2019) found that a positive DVI results in the higher satisfaction of tourists, further leading to positive word-of-mouth recommendation and revisit intention. Thus, the following hypotheses are presented:

In the tourist literature, consumer satisfaction is an assessment made by customers based on their interactions with service providers, and it can be used to forecast future experiences (Crosby et al., 1990). In tourism, satisfaction primarily relates to the judgment of travelers concerning tourist destinations based on their on-site experience. The contentment of travelers is a favorable mental state created by the experience travelers had while traveling (Meng and Uysal, 2008), and the loyalty of tourists to a destination has a strong correlation with tourist satisfaction.

Previous studies have found that satisfaction affects the future behavioral intentions of tourists (Petrick, 2004; Prayag and Ryan, 2012). Consumer loyalty and satisfaction are firmly and inexorably related according to both practitioners and academics (Oliver, 1999). According to Melo et al. (2017), the least satisfied tourists are less likely to return to a destination compared with their more satisfied counterparts. Ragab et al. (2019) suggested that tourist satisfaction positively and significantly influences not only the positive word-of-mouth recommendation but also the revisit intention. Those results are in line with that of Chi and Qu (2008), who found that satisfaction positively influenced destination loyalty. The following hypothesis is therefore presented, and this paper, thus, has proposed an integrative theoretical research model based on above discussion (see Figure 1):

Using the deductive approach, a positivist philosophical paradigm was adopted for the empirical testing of hypotheses. The underlying patterns and causal relationships were identified, and the obtained results were generalized. The outcomes of the testing of the hypotheses contribute to the domain of related knowledge. The selection of positivism was to reduce bias, as the single external reality in a value-free manner was examined objectively. The exploratory nature of this study provides valuable new insights and prospects on the phenomenon under study. For the current study, a survey was conducted to obtain a large sample from a large population (Saunders et al., 2011). The survey was deemed reliable by the respondents, and the obtained data can be standardized and easily compared (Sekaran, 2006). Considering the limitations of resources, a cross-sectional research design was considered.

Kaiping Diaolou (clock towers), and the villages in which they stand, lie within the South Asian tropical climate zone in a hilly area of oriental rice cultivation. The city center of Kaiping is located on the Changjiang River, 140 km (87 miles) west of the Pearl River Delta, on the outskirts of the county of Kaiping. The construction of the Diaolou began in the 14th-15th centuries, but the peak period of its construction was in the late 19th century and the beginning of the 20th century (UNESCO World Heritage Convention, 2007); the Diaolou and the surrounding villages followed traditional cultural and land-use patterns from the turn of the 20th century, notable for the tidy traditional communities in front, with high defensive buildings for flood preservation behind the low-rise village residences, which are separated by narrow lanes, with fish ponds and ancestral halls in front and the tall Diaolou surrounded by bamboo groves behind (Huadong, 2013).

For the present study, a non-probability sampling method was considered. It is difficult for social science studies to acquire an accurate sampling frame or track down potential respondents from a population of interest to address research questions (Saunders et al., 2011). A non-probability sampling method is typically based on the subjective judgment of the researcher (Sekaran, 2006). In this study, a convenience non-probability sampling strategy was adopted. Data were gathered based on two characteristics: (1) geography—that is, domestic respondents visiting Kaiping in Guangdong; and (2) time-based—that is, respondents selected during the Tomb-Sweeping holiday in 2021. The Tomb-Sweeping holiday is a long-standing public holiday in China during which the majority of Chinese people engage in tourism activities. This facilitated data collection and ensured a more representative population. The questionnaire items were translated into Chinese using the back-translation method. A pretest with 30 respondents was undertaken to assure the use and validity of the designed instrument and to prevent any difficulties that could compromise the quality of the gathered data.

Based on Cochran’s Formula, a reasonable sample size of 384 respondents was determined (Wang et al., 2020b). Hair et al. (2010a) noted that sample sizes greater than 200 offer an adequate margin of error, although some researchers recommend that a sample size of at least 200 or between 10 and 20 cases per parameter is appropriate for using structural equation modeling (Kline, 2015; Wang and Zhang, 2021). In total, 448 questionnaires were returned; after eliminating invalid and incomplete questionnaires, 406 usable questionnaires were retained, which exceeded the value of the limited respondents. Furthermore, a latent variable was included in the confirmatory factor analysis (CFA) model by connecting it to observable factors, as Podsakoff et al. (2003) suggested that a common latent factor can examine the common method bias (CMB). The new model was evaluated by standardized regression before comparing it with the original model. Both models were found similar after the comparison. Finally, Harman’s single factor test was undertaken to identify if the CMB affected the results. The results showed an exploratory factor analysis with a single factor accounted for 48.498% of the variance, which is less than the 50% benchmark value. This indicated that CMB was not a substantial issue in this study.

The research instrument adopted for this study was a self-administered questionnaire. The questionnaire had four sections. The first section included the three independent variables: LCP, authenticity, and access to local products. The three items related to LCP were adapted from Gursoy et al. (2010); the four items used to measure authenticity were adapted from Lee et al. (2020); the five items related to access to local products were adapted from Puciato (2020). The second section included DVI and tourist satisfaction. The five items used to assess DVI were adapted from Wang et al. (2021a), and the three items related to tourist satisfaction were adapted from ÇavuĢoğlu et al. (2020). The third section was used to assess the endogenous dependent variable: tourist loyalty, for which three items were adapted from ÇavuĢoğlu et al. (2020). The last section elicited relevant demographic characteristics. All of the aforementioned measurement items were evaluated using a five-point Likert scale, ranging from “strongly disagree” to “strongly agree.” The comprehensive research questionnaire is presented in Appendix A.

Table 1 presents the demographic information for the present study. Of the 406 respondents whose answers were used, approximately 37.4% were male, and 62.6% were female; 67% were aged between 18 and 30, 16% between 31 and 45, followed by those under 18 (11.8%), 4.7% between 46 and 60, and 2 respondents were older than 61. In terms of education, 47% of the respondents have completed a 4-year bachelor’s degree, 34.7% have completed a 3-year diploma, 6.9 and 6.7% have completed high school or a master’s and above, respectively, and 4.7% have completed middle school education. Approximately 38.2% of the respondents reported earning 1,701–3,000 CNY (Chinese Yuan) in monthly income, 20.9% of respondents earned 3,001–4,500 CNY, 15.3% earned 4,501–6,000 CNY, 15% earned below 1,700 CNY, and 10.6% of respondents earned more than 6,001 CNY.

To analyze the reliability of the study, Peterson (1994) recommended using an index form for a Cronbach’s alpha value of 0.7 and above for the acceptable internal reliability. For the convergent validity of the measurement model, the composite reliability (CR) should be more than 0.7, and the average variance extracted (AVE) should be greater than 0.5 (Hair et al., 2010a). The results showed that validity and reliability were achieved (see Table 2).

The CR should be higher than the AVE. To assess the discriminate validity, the maximum shared squared variance (MSV) and the average shared squared variance (ASV) were both considered. The MSV and ASV should both be smaller than the AVE (Hair et al., 2010b). No correlation between each construct exceeded 0.9 (Wang et al., 2021b); so, the validity of the measurement model is established, as shown in Table 3.

Next, the fit of the measurement model was checked. The model fit indices demonstrated that the measurement model was a good fit to the data (χ² = 644.172, degree of freedom (df) = 211, p < 0.001, χ²/df = 3.053, root mean square residual (RMR) = 0.039, goodness of fit (GFI) = 0.874, comparative fit index (CFI) = 0.955, adjusted goodness of fit index (AGFI) = 0.835, parsimony goodness of fit index (PGFI) = 0.668, parsimony normed fit index (PNFI) = 0.779, parsimonious comparative fit index (PCFI) = 0.796, non-normed fit index (NFI) = 0.934, incremental fit index (IFI) = 0.955, tucker lewis index (TLI) = 0.946, relative fit index (RFI) = 0.821, root mean square error of approximation (RMSEA) = 0.071) and met the threshold values.

The following stage was to conduct SEM to validate the hypotheses of the study. The outcomes indicated that the suggested structural model possessed a sufficient level of goodness-of-fit statistics (χ² = 747.801, df = 217, p < 0.001, χ²/df = 3.446, RMR = 0.082, GFI = 0.863, CFI = 0.944, AGFI = 0.826, PGFI = 0.679, PNFI = 0.792, PCFI = 0.81, NFI = 0.924, IFI = 0.945, TLI = 0.935, RFI = 0.911, RMSEA = 0.078). Figure 2 and Table 4 show the details for these findings.

According to Wang et al. (2021a), a path coefficient below 0.1 indicates a small effect, a path coefficient of about 0.3 indicates a moderate effect, and a path coefficient above 0.5 indicates a large effect. The results indicate that LCP positively influenced DVI with major effects (β = 0.486, p < 0.001); thus, H1a was supported. Authenticity and access to local products positively and moderately influenced DVI, with authenticity influencing DVI with a path coefficient of 0.328, p < 0.001, and access to local products influencing DVI with a path coefficient of 0.219, p < 0.001. Hence, H1b and H1c were supported. The results indicate that DVI moderately positively influenced tourist satisfaction with a path coefficient of 0.445, p < 0.001, and tourist loyalty with a path coefficient of 0.255, p < 0.001. Thus, H2a and H2b were supported. Furthermore, tourist satisfaction had a major influence on tourist loyalty (β = 0.582, p < 0.001). Thus, H3 was supported.

Most empirical studies have proposed destination familiarity as a single-dimension construct measured through previous destination visiting experience (Stylidis, 2020) that has a positive correlation with destination image (Chi et al., 2020). Recent studies have shown that familiarity with a tourist destination can be conceptualized as having two dimensions: experiential and informational (Chen and Lin, 2012; Kim et al., 2019). However, Kim et al. (2019) indicated that most previous studies on destination familiarity have only measured experiential familiarity and failed to consider the indirect familiarity (without consumption experience) of consumers with a destination, and this element also needs to be incorporated into the measurement of destination familiarity. Thus, this study proposed that informational familiarity can be measured through LCP, authenticity, and access to local products.

First, researchers have underestimated the role of LCP in the perceptions, satisfaction, intentions, and behaviors of tourists. According to Chili and Ngxongo (2017), community involvement can strengthen the character of most tourism destinations. However, most investigations have proposed that the role of LCP should be in tourism planning, development, and operation in either the initial or the progress stage. However, tourist satisfaction tends to be greater in areas where hosts support and take pride in tourism (Richards and Hall, 2003). The positive DVI of tourists is positively influenced by the performance of local residents, and this translates into satisfaction and loyalty toward the cultural heritage destination. The study outcomes thus offer an alternative perspective for considering the role of LCP in tourism activities.

Second, authenticity and access to local products significantly influenced the DVI of tourists for a particular cultural heritage destination. Cultural heritage sites differ from many other tourism products and services, and authenticity plays an important role in determining whether a particular cultural heritage destination can satisfy the DVI of tourists. Tourists from different parts of the world spend their leisure time visiting these remote places, which are not easily accessible. Easy access to local products significantly improves the DVI of tourists for cultural heritage destinations. Therefore, this study provides a basis for future research to replicate and clarify the motivations and perceptions of tourists, as well as their visiting behaviors regarding cultural heritage sites.

Third, research related to cultural heritage destinations has rarely been characterized for developing countries (e.g., China), and insufficient attention has been paid to how local communities influence tourist cultural heritage destinations. Our results confirmed that LCP, authenticity, and access to local products positively and significantly influenced the DVI, satisfaction, and loyalty of tourists toward cultural heritage destinations. These findings are especially significant for communities in developing countries such as China and Egypt (Ragab et al., 2019), where LCP, authenticity, and access to local products could play a significant role in marketing for cultural heritage tourism.

There are several practical consequences for cultural heritage destinations based on the results of this study. First, this study emphasized the importance of destinations developing deeper marketing strategies with LCP, because the feelings of tourists about visiting a particular heritage destination are highly influenced by LCP. Local communities cannot isolate local people from tourism planning, development, or promotion. The invisible information (i.e., informational familiarity) interchange between tourists and local people highly influences the DVI of tourists, thus, local tourism planners should re-consider establishing a role for local people in tourism activities and developing tourism activities involving local people, such as encouraging locals to engage in routine tourism activities, promoting homestays, and recruiting more locals as publicity commentators to stimulate the sense of LCP and promote local culture to visitors.

Second, because the findings of this study showed that authenticity and access to local products positively influenced DVI, thus, cultural heritage sites should ensure the accuracy of the information published on social media and websites about the destination. Visitors should receive the right information so that they can perceive that they made the right decision to visit the destination. Such web content should transfer positive information to potential visitors, such as how easy it is to visit the place, the competitive price of local products, and an atmosphere that is friendly to all visitors. Overall, cultural heritage destinations should create easy access to accurate information so that potential visitors can develop a positive DVI.

The results of this study also showed that DVI positively and significantly influenced tourist satisfaction and loyalty. Destination image can be classified into affective and cognitive images, so cultural heritage destinations operators should not only continue to preserve natural resources (e.g., climate, landscape, environment), but also pay more attention to developing better human-made resources, such as restaurants, hotels, accommodations, theatre festivals, and folklore, among others. Tourist satisfaction and loyalty toward a cultural heritage destination (i.e., Kaiping) can also be assessed in terms of the historical, social, political, and economic characteristics of a destination. Therefore, on the one hand, cultural heritage destination (e.g., Kaiping) operators should provide better services with local communities for potential tourists, while, on the other hand, they also need to develop strategies to preserve natural resources.

The study sample consisted of only two respondents above the age of 61 and did not reflect the total Chinese population, which means that the results may not be generalizable. This study was carried out with a very limited scope and focused on the Kaiping cultural heritage destination in Guangdong province, China. Thus, the context is only applicable to this area and may vary from other places. A sample size larger than 30 and less than 500 has been deemed appropriate for most research (Wang et al., 2020a), and this study has successfully collected 406 respondents for data analysis, but a larger sample size could provide more normal distribution and produce better outcomes (Saunders et al., 2011). This study is also one of few to evaluate the influence of LCP on the DVI and satisfaction of tourists from the decision-making process perspective of a tourist. The model should be duplicated and tested in other locations to further confirm its usefulness and validity.