From Trust to Triumph: What Drives Tourists to Recommend Thailand’s Hidden Cultural Gems?

Abstract

This study aims to explore the factors influencing word-of-mouth (WOM) intentions among cultural tourists in Thailand’s southern border provinces. This study develops a comprehensive framework predicting WOM intentions by combining the Stimulus–Organism–Response (S-O-R) model, the theory of planned behavior (TPB), and the cognitive–affective–conative (CAC) model, given the critical role of WOM in tourism marketing. Data were collected from 386 cultural tourists through purposive sampling using a self-administered questionnaire. The study employs structural equation modeling (PLS-SEM) to investigate relationships between trust, attitude, satisfaction, and WOM intentions. The findings reveal that trust, satisfaction, and attitude significantly impact WOM intentions. Social influence strongly predicts trust, while the perceived quality of behavior influences attitude. Satisfaction is driven by perceived value, the quality of the experience, the quality of the physical environment, and tourist engagement. This study contributes to cultural tourism literature by integrating different behavioral theories to provide a robust WOM model. This study’s theoretical contributions and practical implications for scholars, policymakers, tourism marketers, and cultural site operators will be further discussed.

1. Introduction

The tourist sector is fundamental to Thailand’s economy, making substantial contributions to government revenue. In recent years, the World Travel and Tourism Council (WTTC) declared that tourism has constituted roughly 15–20% of Thailand’s GDP, becoming one of the nation’s principal economic drivers (WTTC, 2023). Prior to the COVID-19 epidemic, the Tourism Authority of Thailand (TAT) stated that Thailand attracted approximately 40 million international tourists each year (2019), yielding over 1.9 trillion Thai baht (approximately 60 billion USD) in revenue (TAT, 2023). The Bank of Thailand (BOT) detailed that the business sustains millions of employees, both directly and indirectly, spanning sectors like hotels, transportation, and retail (BOT, 2021). The epidemic precipitated a significant downturn in tourism revenue; nevertheless, the sector is rebounding as foreign travel recommences. The Ministry of Tourism and Sports (MOTS) maintains a focus on tourism by implementing programs such as visa waivers, advertising campaigns, and infrastructural enhancements to draw visitors (MOTS, 2023). Tourism is a dynamic source of income for Thailand, playing a significant role in economic growth and employment (Simasathiansophon et al., 2020). The Tourism Authority of Thailand (TAT) reported that the number of foreign tourists visiting Thailand in 2023 was approximately 30 million people, traveling from China, ASEAN countries, India, the United States, and others, which generated 1.8 trillion baht (TAT, 2023).

One of the most exciting tourism sectors, besides natural tourism, is “cultural tourism”. Cultural tourism is tourism that uses the culture and tradition of each country as a product to differentiate itself from other types of tourism, which adds value to tourist destinations (Choeichuenjit & Sapsanguanboon, 2014). Cultural tourism accounts for the largest market size compared to other tourism sectors. Cultural tourism, as a niche market in the service-intensive tourism industry, has distinctive characteristics, experiences, and different individual consequences (Nanggong & Mohammad, 2024). Regarding the global level, the cultural tourism market is valued at approximately USD 900.00 billion, and it is expected to grow by 15% per year from 2014 to 2024, worth USD 3641.00 billion. Hence, it can be concluded that cultural tourism is more attractive due to the substantial global market and high growth rates (UNESCO, 2023). Cultural tourism in Thailand includes historical and religious sites, traditional festivals, local arts and crafts, cultural villages, and experiences. The Thai government has set many initiatives to promote cultural tourism through campaigns like “Amazing Thailand”, which highlights the country’s cultural diversity and heritage (TAT, 2023). Kasamsuk et al. (2018) define cultural tourism’s identity as architectural (chapels, pagodas, and walls), artistic (drawings in temples depicting Buddha’s life and teachings), sculptural (Buddha, angels, and gods), historical (references to Buddha’s life and teachings), and ritual and religious (i.e., Thai traditional dancing during Buddha days). Cultural tourism is important for many reasons, including economic and social benefits, identity creation and reinforcement, image construction, and cultural and historical preservation. It allows peaceful and understanding collaboration (Simasathiansophon et al., 2020).

Word of mouth (WOM) refers to informal communication among individuals who have experienced goods or services, sharing their evaluations (C. C. Li, 2024). WOM involves the transmission of opinions from one consumer to another, serving as a critical phase in the consumer decision-making process that persuades individuals to utilize specific products or services. This communication effectively penetrates the target audience, conveying credible messages that have been shown to influence behaviors and attitudes across various settings, including offices, homes, schools, blogs, and social networking sites (SNSs), wherever consumer discussions occur (Jalilvand et al., 2017). Baloglu and McCleary (1999) assert that WOM exerts a more favorable influence on a customer’s perceived image compared to professional advice, advertisements, and media sources such as books, movies, and news. Research has shown that word of mouth influences 76 percent of all purchasing decisions. Approximately 3.4 billion word-of-mouth conversations occur daily, with 2.3 billion focused on brands (Balter, 2008). Prior studies have sought to elucidate the impact of word of mouth on consumer behavior. Jalilvand and Samiei (2012) demonstrated that word-of-mouth communication can have a positive influence on tourists’ destination choices.

Furthermore, cultural tourism in Thailand significantly influences the decisions of both domestic and international tourists to visit. Cultural tourism destinations typically encompass historical sites, including chapels, pagodas, and walls, as well as museums or galleries, art performance events, and distinctive local lifestyles (Kasamsuk et al., 2018). However, Thai tourism exhibits a lower competitiveness relative to countries such as Japan, China, Myanmar, India, Cambodia, South Korea, and Vietnam. The primary concern is the insufficient organization and data integration of Thai culture among the relevant institutions (Simasathiansophon et al., 2020). Research by Rasoolimanesh et al. (2019) and Alrawadieh et al. (2019) indicates a positive correlation between tourist engagement and both destination loyalty and intentions to revisit. Although existing studies emphasize the important role of engagement in influencing tourist experiences and, in turn, tourism’s development at a destination, the impact of tourism engagement on the behavioral intentions of cultural tourists remains insufficiently explored in the current literature. Likewise, the effects of satisfaction, attitude, and trust on the behavioral intentions of cultural tourists have not adequately been addressed in the extant literature. Therefore, this study aims to address that gap by proposing a model that would depict the specific factors influencing culture tourists’ WOM.

Overall, the literature on cultural tourism is also somewhat limited. Moreover, the previous studies did not go into enough depth to explain tourism culture, especially on the uptake of their behavior preferences. This conclusion underscores the necessity for additional investigation to enhance our comprehension of cultural tourists’ experiences and participation, as well as to further elucidate word-of-mouth dynamics within the cultural tourism context. Consequently, this study examines the impact of many elements on behavioral intentions, specifically focusing on word-of-mouth (WOM) intention, by integrating the Stimulus–Organism–Response (S-O-R) framework, the cognitive–affective–conative model, and the theory of planned behavior (TPB). This study contributes to the existing literature on cultural tourism by creating a comprehensive depiction of the behavioral intentions of cultural tourists. The findings of the present study can aid cultural sites’ marketers and operators in formulating effective marketing strategies for tourism destinations by fostering or enhancing tourist engagement that encourages subsequent behavioral intentions. Hence, the research question that guides this study is, “What are the significant factors that can impact the word-of-mouth intention of cultural tourists?”

2. Literature Review

Word of Mouth (WOM), originating from marketing, is defined as verbal, interpersonal communication regarding a product or service (Lai et al., 2020). Similarly, Jalilvand et al. (2017) defined word of mouth (WOM) as informal communication occurring between consumers and individuals interested in products or services. WOM communication has emerged as a crucial concept in modern marketing (Kankhuni & Ngwira, 2021) due to consumers’ growing resistance to traditional advertising and other marketing communications. Positive word of mouth reflects customer loyalty and enhances the probability of purchasing the company’s products (Zhong et al., 2024). Previous studies suggest that word of mouth significantly impacts tourists’ behaviors and consumption choices (Jalilvand et al., 2017). The WOM concept is grounded in cognitive, emotive, and interactionist perspectives (Taheri et al., 2021). In the marketing literature, consumer engagement has long been thought to indicate the likelihood of WOM intention (S. Chen et al., 2021).

Zhong et al. (2024) explored the incentive needs influencing positive word of mouth (PWOM) among international travelers and analyzed the variation in these needs in relation to individualism and collectivist cultural orientations. Nanggong and Mohammad (2024) examined the essential elements of cultural tourism experiences to predict tourist behaviors, aiming to improve electronic word of mouth (e-WOM) and impact cultural destinations’ image. C. C. Li (2024) analyzed the determinants influencing visitors’ travel choices amid the COVID-19 pandemic. This research analyzed additional factors influencing travel plans, such as perceived risks, awareness, and interpersonal communication. Karim et al. (2024) identified factors affecting passengers’ behavioral intentions regarding beach resorts in Bangladesh. This study examines the moderating effects of the experience and satisfaction a destination offers on the relationship between destination service variables and travelers’ behavioral intentions. Arifah et al. (2024) examined the effect of AR, source credibility, information needs, and the quantity of e-WOM information on information’s usefulness and the effect of information’s usefulness on visiting intentions among TikTok social media users who are looking for information about tourist attractions in the Thousand Islands in Canada. Mathew et al. (2024) found that tourism destination development and visitor satisfaction are significantly enhanced by responsible tourism practices. This study is an effort to comprehend how responsible tourism practices affect visitor satisfaction, re-visit intention, and word of mouth.

Silaban et al. (2023) showed how e-WOM on Instagram affected travel behavior in Indonesia. The necessity for information is the main driver influencing the value of that information. Additionally, the credibility of sources and the quality of their arguments are essential in improving the practical value of information. Guo et al. (2022) examined the key variables influencing the use of automated buses and the intention to engage in word-of-mouth communication in Stockholm, Sweden. Users’ needs and travel demands significantly influence intentions to adopt a new public travel mode and to recommend the service to others. Stylidis and Quintero (2022) found that a place’s image influenced residents’ perceptions of economic, socio-cultural, and environmental impacts, whereas knowledge of tourism affected only the economic impacts. Furthermore, the placement of images and support for tourism influenced residents’ word-of-mouth intentions in Spain. Taheri et al. (2021) established a framework for WOM, identifying tourism factors, quality of service, and perceived value as significant antecedents influencing WOM referrals.

In order to evaluate the impact of e-WOM and personality traits on the decision to visit a certain Vietnamese tourist attraction, Tapanainen et al. (2021) used the Big Five personality traits and the Information Adoption Model (IAM). The results demonstrated that the attributes of agreeableness, extraversion, and imagination—which are components of personality—influenced the adoption of information about tourist locations. Research by Rasoolimanesh et al. (2021) examined word of mouth (WOM) as a result of visitor engagement, finding that a higher level of engagement significantly affects WOM intention. According to Teng (2021), tourists’ involvement was boosted by film tourism experiences, which had a beneficial impact on their behavioral intentions. In order to understand why park visitors binge-eat at Yanchep National Park in Australia, Esfandiar et al. (2021) used a combination of the norm activation model and the theory of planned behavior (TPB). It became clear from this that visitors’ binging behavior is primarily driven by their personal conventions. Hence, park officials should emphasize the importance of visitors’ prominent personal standards. The impact of tourist interaction dimensions on domestic heritage visitors’ intent to return to Kashan, Iran, and spread the word was studied by Seyfi et al. (2021). With the use of Instagram posts made by overland travelers to African places, Kankhuni and Ngwira (2021) studied the relationship between engagement, perceptions of soundscapes, satisfaction, MTEs, and e-WOM. Lai et al. (2020) examined the relationship between perceived trust, membership, and e-WOM intention for sports events. M. Ali et al. (2020) developed a framework for tourist satisfaction, loyalty, and WOM to select local foods in Pakistan.

N. C. Chen et al. (2018) conducted an empirical study in Shanghai, China, and Sydney, Australia, to motivate residents to communicate with potential tourists by investigating local residents’ positive word-of-mouth behavior about their destination region. Jalilvand et al. (2017) examined the antecedents that may influence tourists’ WOM about restaurants in Tehran, Iran. They found that the quality of the food, quality of personal interactions, quality of the physical environment quality, and perceived value influence WOM behavior. Using data collected from domestic visitors to Japanese heritage sites, Bryce et al. (2015) investigated the connections between cultural uniqueness, perception of authenticity, engagement, and the resulting behavioral implications. C.-F. Chen and Chen (2010) found that relationships between the quality of experiences, perceived value, satisfaction, and behavioral intentions appear to be evident in heritage tourism.

Kornsuphkit et al. (2024) investigated the antecedents the intention of revisiting of Thai cultural tourists in Thailand. They revealed that travel motivation, novelty seeking, the destination’s image, and memorable experiences positively affect the satisfaction, WOM, and revisit intentions of Thai cultural tourists in Thailand. Jaipong (2024) determined the relationship between social influence, electronic word of mouth (E-WOM), previous experience, and tour service purchase intentions among Thai tourists abroad. Wang (2024) investigated the influence of Chinese tourists’ data, tourist behavior, and information sources on Chinese tourists’ satisfaction with travel in Thailand. Punpairoj et al. (2023) identified the motivational factors that encourage travelers to adopt augmented reality (AR) applications when exploring tourism destinations and investigated the mediating effects of personal word of mouth (WOM) on the relationships between innovativeness and revisit intention outcomes among Thai travelers. Namahoot et al. (2023) proposed a model that examined innovativeness (i.e., service innovation and technology innovation) to use AR as the mediating role of attitude and word of mouth (WOM) in Thailand’s tourism industry. Cheunkamon et al. (2022) found that the factors of commitment, satisfaction, and trust directly impacted tourist loyalty to traveling via the domestic public transport system. Factors with a significant indirect impact on tourist loyalty are the service quality of tourism logistics and tourism supply chains. Khadiza and Tapachai (2022) found a significant direct relationship between Thai tourists’ attitudes, subjective norms, and perceived behavioral control with their travel intention to Bangladesh and a significant indirect relationship between the destination’s image and intention. Cheunkamon et al. (2020) found that satisfaction, attitudes, subjective norms, and perception can be used to control behavior, with a positive direct effect on domestic tourists’ intentions to use social media for travel planning. Simasathiansophon et al. (2020) indicated that personal factors (i.e., age and career) and tourist behaviors (i.e., provincial destinations, frequency of traveling, the reason for traveling, and accompaniers) affected the tourist’s decision to visit cultural attraction places in the central part of Thailand. Miao (2015) showed that e-WOM significantly affected Chinese tourists’ behavioral intention toward visiting Thailand by affecting their attitude, subjective norms, and perceived behavioral control.

To sum up, there is a lack of research in the literature that investigates what variables impact the WOM intention of cultural tourists visiting Thailand’s southern border provinces (i.e., Pattani, Yala, Narathiwat). As a last point, the literature does not reveal how to investigate WOM intention by combining the Stimulus–Organism–Response (S-O-R) theory, the theory of planned behavior (TPB), and the cognitive–affective–conative (CAC) model. This research adds to the literature by incorporating these three theories to predict WOM intention among travelers. That is to say, the increasing WOM intention among cultural tourists requires extensive research. In particular, studies focusing on the factors that significantly affect cultural tourists’ WOM intention can be used to increase the travel rate to Southern Thailand successfully. Accordingly, this research sets out to address such knowledge gaps.

2.1. Theoretical Background

Based on the premise that people act rationally, the Stimulus–Organism–Response (S-O-R) theory evolved from the input–output model. In the classical S-O-R model, the stimulus is the external factor that has an effect on the individual’s internal states; in other words, it is an influence that stimulates the individual (Eroglu et al., 2001). On the other hand, the organism is the internal process or structure that mediates the relationship between the external stimulus and the person’s actions, reactions, or responses. Activities, including perception, physiology, emotion, and cognition, make up these intermediate processes and structures (Bagozzi, 1986). Customers may choose to approach or avoid a product or service in the end, and the reaction in the S-O-R model stands in for the results (Robert & John, 1982). According to this study, the S-O-R model may be able to explain people’s reactions. It prioritizes aspects related to the events occurring within the organism (see Figure 1). Although the S-O-R theory has changed and become much more adaptable over the years, it was originally primarily applied to in-site retail activity when it first emerged (Eroglu et al., 2001). Recent studies have employed SOR theory in the tourism and hospitality context (Jiang et al., 2024; H. Liu et al., 2024; H. Qiu et al., 2023; Qiuai et al., 2024). For example, Jiang et al. (2024) combined the S-O-R and experience economy theories to discover factors prompting users’ continuous intentions in cultural heritage virtual tourism. H. Liu et al. (2024) developed an integrative model based on the S-O-R framework to investigate destination loyalty.

The theory of planned behavior (TPB) states that the strongest predictor of action is whether a person intends to pursue a goal. Attitude, subjective norms, and perceived behavioral control affect behavior intention (Ajzen, 1991). This theory holds that a character’s power of intention drives their actions. This hypothesis is used to research people’s behavior. Critics like M. Ali et al. (2018b) say users’ action patterns depend on circumstances and demographics. He suggested that the TPB model needs empirical evidence and extensions to be broadly implemented. Pratiwi (2018) discovered that customer attitude and perceived behavior control strongly influence behavioral intentions but subjective norms do not. Based on earlier studies, scientists advised against interpreting the TPB model’s components too quickly. These results depend on the TPB model’s biased behavior intention determinants, and it is unclear how they connect to consumers’ behavior processes. This research updates the TPB model’s determinants by adding culture tourism-relevant components to the prior findings. Applying TPB to culture tourism explains WOM’s intention more fully. One shortcoming of TPB is that it only evaluates cognitive processes and not the individual’s needs or motivations before acting. Recent studies have adopted the TPB theory in the culture tourism context (Tamuliene et al., 2024; Zhang & Ran, 2024).

Furthermore, the cognitive–affective–conative (CAC) model utilizes cognitive, affective, and conative processes. It zeroes in on the connection between conative elements like WOM and cognitive factors like tourist engagement. Thus, Fishbein’s (1967) cognitive–affective–conative paradigm for explaining attitude serves as the conceptual framework for this study. Several prior research studies have employed this model to examine user engagement, behavioral intention, and word-of-mouth communications (Harrigan et al., 2017; Su et al., 2020). Agapito et al. (2013) validated that a conative image is the outcome of a cognitive image by testing and confirming the hierarchical nature of the interaction between the cognitive, affective, and conative components in the destination’s image. In the context of historical tourism, Fu (2019) looked into how cognitive loyalty affected conative loyalty. Also, according to Harrigan et al. (2017) and Koenig-Lewis et al. (2021), the outcome variable of consumer engagement is behavioral intention. Budiatmo and Listyorini (2025) investigated how a person’s perception of a place affects their likelihood of returning there. The behavior of visitors toward returning and recommending was studied by Yang et al. (2022).

In sum, the current research assumes that integrating two or more models could provide even more robustness than those models standing alone (Gillenson & Sherrell, 2002; Sigala et al., 2000). Viewing a discipline from multiple lenses, conceptual frameworks, or theories is argued to be pertinent as it could create valuable insights (Shook et al., 2009). Each model addresses a distinct component of behavioral intentions, where S-O-R captures environmental stimuli, TPB explains cognitive evaluations, and CAC accounts for emotional and conative dimensions. This integration reduces the conceptual overlap. Each model’s boundaries reflect that its effectiveness may vary by context, such as between cultural and mass tourism, and that it focuses on pre-behavioral intentions rather than actual behaviors. Recent studies by Jiang et al. (2024) and Zhang and Ran (2024) demonstrate the relevance of combining these models in tourism research. Hence, the literature reviewed lacks knowledge about the integrated S-O-R, TPB, and cognitive–affective–conative model to examine culture tourists’ WOM intention. In light of the current research, this study adds to it by suggesting a model that combines the S-O-R, TPB, and CAC theories to predict individuals’ WOM intention in the culture tourism setting.

2.2. Conceptual Framework and Hypotheses Development

By combining the S-O-R, TPB, and cognitive–affective–conative models, this study builds the research model. An individual’s mental drive to act is what drives WOM’s intention. A method for forecasting how likely culture tourists are to express their WOM intentions is based on their current availability to do so. Figure 1 portrays the proposed research framework.

2.3. Hypotheses Development

Social influence is the effect on travelers’ impressions and choices of recommendations and opinions from friends, relatives, and social networks. Seeing and copying the actions of their social groups helps people to build trust (Boto-García & Baños-Pino, 2022). Developing trust in a place depends very much on social influence. For visitors, trust is essential, since it lowers the apparent risk and uncertainty, thus increasing their likelihood of visiting and recommending the place. Recent research has repeatedly shown that trust is much influenced by social influence (Jalilvand et al., 2017; Lai et al., 2020; Seger-Guttmann & Gilboa, 2023). For example, Jalilvand et al. (2017) discovered that trust in the restaurant sector is influenced by social influence, implying that suggestions from social networks are vital in determining how guests view and make judgments. In a similar vein, Lai et al. (2020) underlined how trust in cultural tourism is mainly driven by social influence. Thus, the following hypothesis is put forward:

H1. 

Social influence positively influences trust.

Traveler satisfaction is significantly influenced by perceived value. Those who value their trip experiences highly are more likely to be happy, since they believe the advantages exceed the expenses. Recent studies repeatedly underline how important perceived value is in raising satisfaction (D. Chen, 2023; Lee et al., 2016; N. Qiu et al., 2024). In cultural tourism, for instance, C.-F. Chen and Chen (2010) revealed that perceived value significantly affects satisfaction, implying that visitors are more content when they believe they are getting good value for money. Comparably, Wu and Li (2017) underlined that in cultural tourism, contentment is mostly driven by perceived value. Consequently, the following theory is put forward:

H2. 

Perceived value has a significant positive influence on satisfaction.

The engagement of tourists in cultural events and interactions with the local population is known as tourist engagement (Seyfi et al., 2021). Because they feel more linked to the location, engaged visitors are more likely to be happy with their trip (Teng, 2021). Recent studies have repeatedly underlined the need for visitor participation in improving satisfaction (Patterson et al., 2006; Rasoolimanesh et al., 2019; Zhou & Yu, 2022). For instance, Patterson et al. (2006) discovered that in cultural tourism, tourist involvement greatly affects satisfaction, implying that active participation in cultural events improves tourists’ emotional and cognitive reactions. In historical tourism, similarly, Rasoolimanesh et al. (2019) underlined that a major factor influencing satisfaction is visitor involvement. Consequently, the following theory is put forward:

H3. 

Tourist engagement positively impacts satisfaction.

The physical surroundings of a destination, including its facilities, cleanliness, and beauty, are known as the quality of its physical environment. Clean and aesthetically appealing surroundings improve the whole experience and increase the likelihood of visitor satisfaction. A physical environment’s quality clearly influences satisfaction, according to recent studies (F. Ali et al., 2016; Fang et al., 2024; Jalilvand et al., 2017; Renganathan et al., 2024). For example, Jalilvand et al. (2017) discovered that satisfaction in the restaurant sector is much influenced by the quality of the physical environment, implying that it shapes visitors’ impressions and experiences. In line with this finding, Fang et al. (2024) underlined how well-maintained physical surroundings improve the whole travel experience. Consequently, the following theory is suggested:

H4. 

The quality of the physical environment has a significant positive impact on satisfaction.

Determining tourist satisfaction mainly relies on the quality of their experience. Excellent experiences marked by enjoyable and memorable activities help to shape a good assessment of the trip. Recent studies have repeatedly shown how the quality of experience shapes pleasure (Domínguez-Quintero et al., 2020; Hsieh & Chuang, 2020). For instance, as it directly affects tourists’ emotional and cognitive reactions, González-Rodríguez et al. (2020) showed that quality of experience greatly affects satisfaction in cultural tourism. Analogously, Hsieh and Chuang (2020) underlined that since it directly affects tourists’ emotional and cognitive reactions, quality of experience is a major factor determining satisfaction in cultural tourism. Thus, the following hypothesis is put forward:

H5. 

The quality of the experience has a significant positive impact on satisfaction.

The perceived quality of behavior—that of local residents and service providers—plays a major influence on visitors’ impressions of a place. Positive interactions, helpfulness, and friendliness help to create a good attitude that can improve general satisfaction and a chance to recommend the place (Pandža Bajs, 2015). Recent research repeatedly reveals that attitudes and satisfaction are much influenced by the perceived quality of behavior (Karim et al., 2024; Yasa et al., 2022). For example, Lee et al. (2016) observed that in cultural tourism, visitors’ opinions and satisfaction are influenced by the perceived quality of behavior. Likewise, Nam et al. (2016) underlined how perceived behavior of a good quality results in more positive sentiments, thereby impacting the customer decision-making process. Consequently, the following theory is put forward:

H6. 

The perceived quality of behavior positively influences attitude.

WOM activity is highly influenced by trust. Those who trust a place are more likely to share good experiences and suggest it to others. Recent studies repeatedly reveal that WOM behavior is much influenced by trust (Jalilvand et al., 2017; Lai et al., 2020; J. Liu et al., 2019). For example, Jalilvand et al. (2017) revealed that WOM behavior is much influenced by trust, implying that visitors who trust a location are more likely to propose it. In line with this finding, J. Liu et al. (2019) underlined that good WOM suggestions are driven mainly by trust. Thus, the following hypothesis is put forward:

H7. 

Trust has a significant positive impact on WOM.

WOM behavior is strongly influenced by satisfaction. Satisfied visitors are more likely to tell others about their good experiences and suggest the location (Rasoolimanesh et al., 2022). Recent studies repeatedly show how important pleasure is in influencing WOM behavior (Hossain et al., 2024; Ng et al., 2023). For instance, research by Santos and Meléndez (2016) indicates that WOM behavior in cultural tourism is much influenced by pleasure, implying that happy visitors are more likely to propose a location to others. Comparatively, Jalilvand et al. (2017) underlined that good WOM suggestions mostly depend on satisfaction. Thus, the following hypothesis is put forward:

H8. 

Satisfaction has a significant positive impact on WOM.

As the theory of planned behavior (Ajzen, 1991) suggests, attitude is a major driver of behavioral intentions. Positive WOM recommendations most certainly follow from a good attitude about a destination. Attitude clearly affects WOM behavior in travel environments, according to recent studies (Hasan et al., 2023; Porras-Bueno et al., 2023). For example, Kamata (2022) discovered that those who have positive opinions of a location are more likely to suggest it to others. Comparably, Shome (2021) underlined that WOM conduct is shaped in part by attitude. Positive opinions of a place or service inspire visitors to share their experiences with others (Rasoolimanesh et al., 2021). Consequently, the hypothesis that follows is put forward:

H9. 

Attitude has a significant positive impact on WOM.

3. Research Methodology

The research setting was in Thailand’s southern border provinces (i.e., Pattani, Yala, Narathiwat). Culturally integrated tourism is being promoted across the three southern border provinces as part of the government’s smart city “Triangle of Stability, Prosperity, and Sustainability” plan (SOUTH, 2017). In addition, the three southern border provinces have many cultural and ecotourism places worth visiting for their historical and cultural importance and outstanding natural beauty (Miao, 2015). Thais in the south are cosmopolitan and have their own culture. The government wants to boost cultural and ecotourism in the Thai South to highlight its distinctive charms. One example is Narathiwat’s Tak Bai district’s historic Buddhist shrine, also known as “Wat Phithak Phaen Din Thai”. Murals depicting the local culture and lifestyle and beautiful Southern Thai Buddhist buildings are there. The Lim Ko Neaw Goddess Shrine and Krue Se (Kersih) Mosque are two of Pattani’s many attractions. Visitors go to Yala’s Wat Khuha Phimuk cave temple, which includes an ancient reclining Buddha. The Betong district is known for its tourist attractions. In Yala, Aiyoe Weng’s sea of clouds is a new attraction (SOUTH, 2017).

3.1. Sampling Technique, Size, and Data Collection

This study adheres to ethical principles, including voluntary participation, informed approval, confidentiality, and privacy. The unit of analysis was cultural tourists of Thailand’s southern border provinces, since they represent the sample frame best. Data for the current research came from a survey that was administered in 2024; it followed a quantitative research method. A purposive sample method was used to pick the participants. This study adopted the self-administered questionnaire technique to collect the data. With this technique, the respondents could read the survey questions and record their honest answers freely, minimizing the tendency for social desirability bias to occur (Sekaran & Bougie, 2020). The minimum sample size was obtained using the G*Power tool (Faul et al., 2009). A medium effect size of 0.15, an error type of 0.05, a power of 0.95, and nine predictors are all part of the G*Power parameters. Because of this, the researchers concluded that 166 cases were the bare minimum for a representative sample. Only 398 questionnaires were successfully collected out of the 500 distributed. However, 12 of them were useless due to being unfilled (i.e., large sections of the questionnaire’s items were incomplete or had a repetitive number on all Likert scale items). To conclude, the ultimate usable sample was 386 responses, giving a response rate of 77.2 percent. Next, each variable’s missing data were analyzed. Data screening showed fewer than 5% missing data. Howell (2007) stated that missing data of up to 10% may not affect the interpretation of the findings. This study used IBM SPSS’s Expectation–Maximization (EM) algorithm to resolve missing data. Structural equation modeling was used with SmartPLS 4.0 software (Ringle et al., 2022) to analyze the direct and indirect interactions between the research variables. Descriptive data were also analyzed using IBM SPSS 26.

3.2. Variables Instruments

A thorough literature review informed this research’s instrument’s combination of validated measurements. The validated measures were slightly changed to fit the research population. Four academics from the tourism and hospitality disciplines and one tourist guide pre-tested the first study instrument. This was to verify the preliminary scales’ reliability and validity and check unidimensional ideas (Cooper & Schindler, 2003). The questionnaire items underwent testing in a pilot study to assess the scale’s reliability, employing the Cronbach alpha statistic. The results demonstrated that all the items exhibited reliability, as their values exceeded the threshold of 0.70 (Tavakol & Dennick, 2011). The initial portion of the survey was designed to gather demographic information from the participants. In Section 2, we measured the conceptual model’s constructs. We used a “7-point Likert scale” to quantify these variables. Items to measure PV, quality of physical environment, trust, satisfaction, and WOM were adopted from Jalilvand et al. (2017). The items used to assess quality of experience were taken from the works of C.-F. Chen and Chen (2010). We used items from Esfandiar et al. (2021) to measure social influence, perceived behavioral control, and attitude. Teng (2021) was referred to for the items of tourist engagement. Additionally, we selected three items adapted from Lin et al. (2015) as a marker variable that would be collected in the same survey, but they were not included in the model being tested: (1) “once I have come to a conclusion, I am not likely to change my mind”; (2) “I do not change my mind easily”; and (3) “my views are very consistent over time”.

3.3. Data Analysis

Using SmartPLS 4.0 and partial least squares (PLS) analysis, this study assessed the suggested research framework (Ringle et al., 2022). Analytical methods were classified by Anderson and Gerbing (1988) into two categories: testing postulated relationships and outer model assessment. Numerous recent studies models (Alam et al., 2022; Cortez et al., 2024; Guillén et al., 2022; Jaboob et al., 2024; Sitar-Taut & Mican, 2021) have made use of PLS-SEM for analyzing research models. The primary advantage of PLS-SEM is this discovery. In order to come up with theories, PLS-SEM examines causal–predictive model linkages (Chin et al., 2020). According to Y. Liu et al. (2022), maximizing the variation in a target construct makes it more predictable. According to Becker et al. (2023), Sharma et al. (2022), and Sarstedt et al. (2022), their approach outperforms factor-based SEM in terms of statistical power and successfully bridges the explanation–prediction gap. Scores for all indicators can be examined using ordinary least squares regressions (J. Hair & Alamer, 2022). For these reasons, PLS stands out as the ideal approach for this research.

3.4. Common-Method Variance (CMV)

PLS-SEM analysis does not consider CMV to be a major concern. The importance of CMV in PLS analysis is a matter of debate (F. Ali et al., 2018a; Ghasemy et al., 2020). According to Kock (2015), the VIF values of the inner model are less than 3.3, which means that the current model is free of CMV. At several levels, starting with study design and data collecting, it is possible to reduce the effects of CMV that arise from similarities in data-gathering procedures. This research looks at the statistical and procedural fixes made before and after data collection. So, following the work of Rönkkö and Ylitalo (2011), we used a marker variable approach to investigate the CMV problem. The PLS marker model and the baseline model are compared in this work (Miller & Simmering, 2023). With R2 values below 10%, we find small changes in word of mouth (2.7%), satisfaction (0.4%), trust (9.6%), and attitude (7.9%). It follows that CMV poses little threat to this study’s methodology (Soliman et al., 2024).

3.5. Profile of Respondents

Table 1. Demographic Information.

Respondents’ Profile	Frequency	Percentage
Age
18–25 years old	55	14.2%
26–35	92	23.8%
36–45	139	36.0%
46–55	83	21.5%
>55	17	4.5%
Gender
Male	205	53.1%
Female	181	46.9%
Provinces visiting
Pattani	215	55.7%
Yala	109	28.2%
Narathiwat	62	16.1%
Local or non-local tourist
Local	204	52.8%
Non-local	182	47.2%
Source of destination information
Advertisement	24	6.2%
Family and friends	197	51.0%
Social media	161	41.7%
Others	4	1.1%
Visiting group
Alone	25	6.5%
With kids	14	3.6%
With friends	119	31.0%
With family	216	56.0%
With an organizer	12	2.9%

describes most respondents as males (53.1 percent), and the most common category is ages between 36 and 45 years old. Likewise, local tourists are dominant (52.8 percent). The touristic places mostly came from Pattani province (55.7 percent). Most tourists obtained their information about the destinations to visit from their families and friends (51%), then from social media (47.7%). They came to Southern Thailand with their families (56%) and friends (31%).

4. Research Findings

4.1. Measurement Model Assessment

J. Hair and Alamer (2022) and Shmueli et al. (2019) emphasize the necessity of evaluating the outer models in the process of developing the proposed model. The outer model, as detailed in

Table 2. Factor loading, CR, and AVE.

Constructs	Items	Loadings	CR	AVE
Attitude	Att.1	0.744	0.841	0.570
	Att.2	0.763
	Att.3	0.740
	Att.4	0.773
Experience Quality	ExQ.1	0.737	0.814	0.593
	ExQ.2	0.790
	ExQ.3	0.787
Perceived Behavior Quality	PBQ.2	0.633	0.752	0.506
	PBQ.4	0.670
	PBQ.5	0.818
Physical Environment Quality	PEnQ.1	0.719	0.806	0.510
	PEnQ.2	0.758
	PEnQ.3	0.686
	PEnQ.5	0.691
Perceived Value	PV.1	0.770	0.833	0.625
	PV.2	0.839
	PV.3	0.760
Social Influence	SI.1	0.722	0.806	0.509
	SI.2	0.736
	SI.3	0.690
	SI.4	0.706
Satisfaction	Sat.1	0.763	0.812	0.590
	Sat.2	0.771
	Sat.3	0.770
Tourist Engagement	T_Eng.1	0.602	0.898	0.504
	T_Eng.2	0.647
	T_Eng.3	0.700
	T_Eng.4	0.634
	T_Eng.5	0.614
	T_Eng.7	0.671
	T_Eng.8	0.645
	T_Eng.9	0.661
	T_Eng.10	0.633
	T_Eng.11	0.678
	T_Eng.12	0.577
	T_Eng.13	0.603
	T_Eng.15	0.583
Trust	Tut.1	0.694	0.830	0.549
	Tut.2	0.762
	Tut.3	0.770
	Tut.4	0.736
WOM	Wom.1	0.735	0.830	0.505
	Wom.2	0.716
	Wom.3	0.729
	Wom.4	0.727
	Wom.5	0.604

Note. Items of PBQ.1, PBQ.3, PEnQ.4, T_Eng.6, and T_Eng.14 have been removed.

, is evaluated using the average variance extracted (AVE), discriminant validity, composite reliability (CR), and factor loading metrics. PBQ.1, PBQ.3, PEnQ.4, T_Eng.6, and T_Eng.14 with low loadings were subsequently dropped to obtain a better AVE. They were removed due to low loadings and their overlap with other items capturing similar aspects of service provider interactions, ensuring the construct’s unidimensionality. These exclusions did not compromise the conceptual scope of the constructs, as the remaining items adequately captured the intended dimensions (e.g., tourist engagement’s multifaceted nature) (J. F. Hair et al., 2019). The AVE must be greater than 0.5, which will reflect at least 50% of the items, and the CR must be greater than 0.7 (J. F. Hair et al., 2019). However, other low loadings were kept, since their values were between 0.40 and 0.708. They can be retained once the measures of AVE and CR have already met the thresholds (J. Hair & Alamer, 2022).

4.2. Discriminant Validity

Table 3. Discriminant validity (HTMT).

	Attitude	Experience Quality	Perceived Behavior Quality	Perceived Value	Physical Environment Quality	Satisfaction	Social Influence	Tourist Engagement	Trust	WOM
Attitude
Experience Quality	0.799
Perceived Behavior Quality	0.643	0.656
Perceived Value	0.859	0.744	0.559
Physical Environment Quality	0.740	0.899	0.571	0.729
Satisfaction	0.835	0.822	0.892	0.827	0.839
Social Influence	0.735	0.718	0.818	0.686	0.678	0.650
Tourist Engagement	0.813	0.899	0.678	0.825	0.810	0.834	0.808
Trust	0.869	0.838	0.723	0.828	0.728	0.754	0.861	0.898
WOM	0.804	0.820	0.719	0.885	0.731	0.869	0.744	0.837	0.838

displays discriminant measures using the HTMT procedure. To evaluate discriminant validity, the approach proposed by Henseler et al. (2015) was utilized. The discriminant validity was assessed in this study utilizing the methodology proposed by Ramayah et al. (2017). The HTMT values are below 0.90, indicating that discriminant validity is not an issue, as stated by Gold et al. (2001).

5. Structural Model Assessment

The analysis of the internal model constitutes the subsequent phase. This research examines the inner model through assessments of multicollinearity, R², PLS_predict, path coefficients, and model fit metrics (J. Hair & Alamer, 2022).

5.1. Multicollinearity Test (VIF)

Prior to conducting the assessment, it is crucial to validate the inner model to ensure that collinearity does not undermine the integrity of the regression results (J. F. Hair et al., 2019). The partial regression scores of the latent variables linked to the predictor constructs produce a variance inflation factor (VIF). Multicollinearity arises when correlated predictors produce redundant information. Value was determined through the use of VIF. A low value of VIF suggests that the variables exhibit a minimal correlation with one another. A VIF value that is less than 3.3 is deemed acceptable (Kock, 2015). All VIF values are below 3.3 according to the research model. This signifies that multicollinearity is not present in the research model.

5.2. Coefficient of Determination (R²)

The impact level for R² may range from 0 to 1. Cohen (1988) classified R² values for endogenous variables as follows: 0.26 is considered substantial, 0.13 is regarded as fairly strong (moderate), and 0.02 is identified as weak.

Table 4. Coefficient of determination.

Endogenous Variables	R Square
Attitude	0.205
Satisfaction	0.499
Trust	0.377
WOM	0.584

shows a substantial level of predictive accuracy for trust, satisfaction, and WOM, while there is a moderate predictive accuracy for attitude. Together, PV, tourist engagement, quality of physical environment, and quality of experience explained 49.9% of the variation in satisfaction. A total of 37.7 percent of the variance in trust was reported by the social influence predictor. Likewise, 20.5 percent of the variance in perceived behavior control was accounted for by the attitude predictor. Finally, 58.4% of the variation in WOM was explained by trust, attitude, and satisfaction.

5.3. PLS Predictive Power

The model’s predictive capacity was assessed for endogenous constructs, which include attitude, contentment, trust, and word of mouth, utilizing tenfold iterations of the PLS_predict technique (Sharma et al., 2022). Initially, we verified that Q²_predict was greater than zero. This scenario suggests that the PLS path model has demonstrated superior performance compared to the average values of the training data indicators. The root mean square error (RMSE) of a linear benchmark model was evaluated against the estimates derived from partial least squares structural equation modeling (PLS-SEM) as outlined by (J. F. Hair et al., 2019). The results of PLS_predict are displayed in

Table 5. Predictive power assessment using PLS_predict.

Items	Q2predict	PLS RMSE	LM RMSE	PLS-LM
Att.1	0.094	1.839	1.711	0.128
Att.2	0.135	1.788	1.792	−0.004
Att.3	0.114	1.866	1.844	0.022
Att.4	0.092	1.916	1.674	0.242
Sat.1	0.276	1.647	0.000	1.647
Sat.2	0.316	1.591	1.721	−0.130
Sat.3	0.274	1.558	1.688	−0.130
Tut.1	0.179	1.761	1.869	−0.108
Tut.2	0.196	1.829	1.786	0.043
Tut.3	0.212	1.829	1.860	−0.031
Tut.4	0.218	1.840	1.903	−0.063
Wom.2	0.215	1.714	1.715	−0.001
Wom.3	0.225	1.792	1.922	−0.130
Wom.4	0.246	1.728	1.741	−0.013
Wom.5	0.195	1.698	1.717	−0.019
Wom.1	0.216	1.764	1.819	−0.055

. All indicators of the endogenous constructs display positive Q²_predict values; there are no negative values present. Additionally, in comparison to the linear model (LM), PLS-SEM demonstrated consistently lower RMSE values. The aggregated findings demonstrate that the PLS route model exhibits a moderate level of predictive capability concerning the target endogenous constructs since the majority of all item differences (PLS-LM) is lower (Shmueli et al., 2019).

5.4. Path Coefficient

The proposed research model’s path coefficients were analyzed by estimating t-values through 10,000 resamples utilizing the bootstrap method (Henseler et al., 2009). The path coefficient may range from −1 to +1. A strong positive association occurs when the value approaches 1, while a negative relationship is indicated when it nears −1. The path coefficients must exhibit a significance level of no less than 0.05. At the 10%, 5%, and 1% significance levels, the critical values for a one-tailed test are 1.645, 1.96, and 2.33, respectively (Ramayah et al., 2014). The nine hypotheses proposed for the study were all completely significant (see

Table 6. Path coefficient and hypothesis test result.

No.	Relationship	Std. Beta	Std Error	t-Value	p Values	f2	Effect Size	VIF	Decision
H1	Social Influence -> Trust	0.614	0.04	16.18	0.00	0.61	large	1.00	supported
H2	Perceived Value -> Satisfaction	0.190	0.07	2.70	0.00	0.03	small	2.11	supported
H3	Tourist Engagement -> Satisfaction	0.208	0.08	2.67	0.00	0.03	small	3.15	supported
H4	Physical Environment Quality -> Satisfaction	0.180	0.07	2.66	0.00	0.04	small	1.82	supported
H5	Experience Quality -> Satisfaction	0.262	0.07	4.01	0.00	0.07	small	2.07	supported
H6	Perceived Behavior Quality -> Attitude	0.453	0.05	10.14	0.00	0.26	large	1.00	supported
H7	Trust -> WOM	0.355	0.06	5.52	0.00	0.14	medium	2.10	supported
H8	Satisfaction -> WOM	0.270	0.06	4.69	0.00	0.11	medium	1.57	supported
H9	Attitude -> WOM	0.264	0.07	3.96	0.00	0.07	small	2.34	supported

). Cohen’s effect size estimation was used to calculate the effect size (Cohen, 1988). If the values are 0.01, 0.09, or 0.25, the effect sizes are categorized as small, medium, and large, respectively. Both social influence and perceived quality of behavior exhibit a sizable effect on attitude and trust, respectively, as shown in Table 6. Likewise, the effect size of satisfaction and trust on WOM is medium. However, the effect sizes of the other variables are small.

5.5. Model Fit and Goodness of Fit (GoF)

The initial step involved assessing the model’s accuracy through a detailed examination of its relationship with two adjustment variables: the normed fit index (NFI) and the standardized root mean square residual (SRMR) (see

Table 7. Testing model fit.

Parameter	Saturated Model	Estimated Model
SRMR	0.061	0.066
NFI	0.918	0.912

). A well-fitting model is characterized by an SRMR value, which is calculated from the discrepancy between the observed correlation matrix and the model-implied correlation matrix, resulting in a value lower than 0.08 (Hu & Bentler, 1998). Henseler et al. (2014) presented the SRMR as a metric for assessing goodness of fit aimed at reducing model misspecification in PLS-SEM. The second metric utilized for assessing a model’s adequacy is the normed fit index (NFI). This incremental fit metric evaluates the chi-square value of the proposed model against a relevant benchmark, offering a supplementary assessment of the model’s fitness (Ramayah et al., 2017). A Non-Normed Fit Index (NFI) value exceeding 0.90 typically signifies a favorable fit (Bentler & Bonett, 1980). The results derived from the estimation of the structural model and its alignment with the measurement model exhibited a notable similarity, attributed to the absence of free paths in the saturated measurement model. The data demonstrate a strong alignment with the model, evidenced by an SRMR value of 0.066, which is below the acceptable threshold of 0.08, and an NFI of 0.912, which exceeds the benchmark of 0.90. Furthermore, the GoF demonstrates the thorough alignment of the model. Tenenhaus et al. (2005) introduced a goodness-of-fit measure, which is calculated as the square root of the product of the mean of R² and the mean of AVE. Wetzels et al. (2009) indicate that a value of 0.360 is appropriate. The GoF value for the proposed model is 0.539, which suggests that the model demonstrates a satisfactory level of fit. Figure 2 presents the results derived from the examination conducted in the final research model.

6. Discussion

This study’s findings offer substantial insights into the determinants of tourist satisfaction and word-of-mouth (WOM) recommendations within the realm of cultural tourism in Thailand. The findings corroborate all the proposed hypotheses, emphasizing the significance of constructs including perceived value, quality of experience, quality of physical environment, perceived quality of behavior, tourist engagement, social influence, trust, and attitude in influencing tourist satisfaction and word-of-mouth behavior. This discussion section elucidates these findings, their consequences, and their congruence with the contemporary literature.

Trust was found to be highly influenced by social context (β = 0.614, p < 0.01), so supporting H1. This suggests that trust in a place is very much developed by the suggestions and opinions of friends, relatives, and social media. This result emphasizes the need to use social media and internet reviews to develop and preserve confidence in cultural tourist sites. For visitors, trust is essential, since it lowers the apparent risk and uncertainty, thus increasing their likelihood of visiting and recommending the place. This result is consistent with Jalilvand et al. (2017), Lai et al. (2020), and Seger-Guttmann and Gilboa (2023). People come to trust people by seeing and copying their actions. In tourism, primarily where visitors rely on social influences to create trust impressions, this is especially pertinent (Boto-García & Baños-Pino, 2022). Supporting H2, the data validated a noteworthy relationship between perceived value and satisfaction (β = 0.190, p < 0.01). This finding aligns with D. Chen (2023), Wu and Li (2017), and C.-F. Chen and Chen (2010), indicating that tourists who perceive a high value are more satisfied. The perception of value shapes expectations and assessments of the whole service experience, thus affecting satisfaction. Consumers who feel they have good value for money are more likely to be happy and participate in constructive activities, including WOM recommendations (Jalilvand et al., 2017). Those who value their trip experiences highly are more likely to be happy, since they believe the advantages exceed the expense. This emphasizes the need for value-for-money experiences for tourist providers, which can be attained by means of competitive pricing, premium services, and distinctive cultural presents.

Moreover, satisfaction was much influenced by tourist interaction (β = 0.208, p < 0.01), thereby supporting H3. This is consistent with studies by Patterson et al. (2006), and Rasoolimanesh et al. (2019), who maintained that tourist engagement raised satisfaction. This implies that the impact of engagement on satisfaction varies depending on the sector and the kind of travel destination. This emphasizes the need to generate chances for an active engagement in cultural tourism. Local celebrations, cultural events, and conventional craft workshops all help to improve engagement and add to a more fulfilling experience. Tourist satisfaction is substantially influenced by the quality of the physical environment (H4: β = 0.180, p < 0.01). This result conforms to recent research stressing the need for the physical environment in forming visitor experiences (Fang et al., 2024; Jalilvand et al., 2017). Maintaining clean and aesthetically pleasant surroundings improves the whole experience and increases the likelihood of visitors’ satisfaction. Maintaining the cleanliness and preservation of cultural monuments, together with offering sufficient facilities, is absolutely vital in Thailand to guarantee a good travel experience.

Furthermore, satisfaction (β = 0.262, p < 0.01) was shown to be very much correlated with quality of experience, so supporting H5. This is in line with research by Rahman et al. (2023) and Domínguez-Quintero et al. (2020), who discovered that high-quality events increase travel and hotel satisfaction. Quality of experience affects several aspects, including service interactions, ambiance, and emotional engagement, all of which, when taken together, influence satisfaction levels (Adhikari & Bhattacharya, 2016). This deduction could entail well-planned cultural trips, real-time local contacts, and immersive events that let visitors actively connect with Thai culture (Kornsuphkit et al., 2024). These findings emphasize the need to provide tourists with excellent experiences to raise their satisfaction. Supporting H6, this study revealed a considerable positive association between attitude (β = 0.453, p < 0.01) and perceived quality of behavior. This result supports earlier studies showing how individuals’ cognitive and affective reactions are shaped by the perceived quality of behavior (Ghorbanzadeh et al., 2021; Silva et al., 2024). According to the Expectancy–Value Theory (Fishbein & Ajzen, 1975), people create attitudes depending on their opinions about the qualities and expected results of an event. Consumers who experience an excellent quality of behavior are more likely to form positive opinions of a good or service (Y. Li et al., 2021; Passafaro, 2020). This result underlines the need for training and awareness campaigns for residents and service providers to guarantee favorable interactions with tourists.

Additionally, WOM was found to be much influenced by trust (β = 0.355, p = 0.01), so supporting H7. This result aligns with Jalilvand et al. (2017) and J. Liu et al. (2019). Trust builds consumers’ confidence in a business or location, which causes them to forward good opinions to others (Lai et al., 2020). However, our results contradict those of Gogoi (2021), who discovered that trust had no direct impact on positive WOM. The encouragement of WOM recommendations depends on developing trust via consistent, dependable service delivery, open communication, and enjoyable travel experiences. Supporting H8, this study validated a notable correlation between satisfaction and WOM (β = 0.270, p < 0.01). This is consistent with research by (Jalilvand et al., 2017), Kankhuni and Ngwira (2021), Ng et al. (2023), Hossain et al. (2024), and Santos and Meléndez (2016) showing that happy consumers are more likely to participate in positive WOM. Satisfaction strengthens customer loyalty and advocacy, leading to increased recommendations to others (Rasoolimanesh et al., 2022). Fostering good WOM depends on ensuring high degrees of satisfaction through value-for-money offers, quality experiences, and good contacts. Finally, WOM behavior was shown to be much influenced by attitude (β = 0.264, p < 0.001). Visitors who are positively inclined to a destination are more likely to tell others about it. This result corresponds with the theory of planned behavior, according to which attitudes affect behavioral intentions (Ajzen, 1991). Also, in line with Kamata (2022) and Shome (2021), this result encourages visitors to share their experiences, and this suggests that the location depends on developing a good attitude through entertaining and unforgettable events. Recent research has also shown how attitude shapes WOM behavior (Rambocas & Metivier, 2024; Sweeney et al., 2020). For instance, Rasoolimanesh et al. (2021) exposed that attitude greatly influences WOM behavior in cultural tourism, implying that a good attitude towards a location increases the possibility of recommending it to others.

7. Conclusions

In the culture tourism setting, cultural tourists’ word-of-mouth intentions have already left their mark. Conversely, there is a dearth of information on what motivates cultural tourists to positively recommend touristic places to others, both in Thailand and elsewhere. Therefore, the purpose of this research was to examine factors affecting cultural tourists’ word-of-mouth intention when visiting Thailand’s southern border provinces. The findings of this investigation intend to boost the effects of cultural tourism in that area. However, there is a dearth of research on that topic. No one has yet addressed word-of-mouth intention by utilizing PLS-SEM analysis in conjunction with the theories of the S-O-R model, TPB theory, and cognitive–affective–conative model. Consequently, it was necessary to study the factors that significantly affect cultural tourists’ word-of-mouth intention. As a result, by developing and testing an empirical model that combines the S-O-R, TPB, and cognitive–affective–conative theories, this research adds to the existing body of knowledge. It extends the existing WOM literature by capturing both external (S-O-R) and internal (TPB/CAC) drivers, unlike prior single-theory studies. Using a questionnaire, we solicited responses from cultural tourists during their visit to Thailand’s southern border provinces. The PLS-SEM method was employed to analyze the data.

Three key factors that cultural tourists considered when deciding whether or not to express their word-of-mouth intention positively were trust, attitude, and satisfaction. This study revealed that social influence had significance on the likelihood of their trust. Likewise, this study found that perceived behavior control was a significant predictor of attitude. Predictors of PV, tourist engagement, quality of physical environment, and quality of experience affect their satisfaction. Moreover, the findings uncover that social influence and perceived quality of behavior have large effects sizes on attitude and trust, respectively. Trust and satisfaction have medium effect sizes on WOM. In sum, this study adds to the overall context of culture tourism thanks to its findings. These findings make this study unique because of how they pertain to discussions about promoting cultural tourism in Southern Thailand. Academics, policymakers, entrepreneurs, and industry professionals involved in culture tourism will find this study intriguing. Thailand can keep becoming a top cultural tourism destination by concentrating on these important perspectives.

8. Research Implications

8.1. Practical Contributions

The study findings provide a basis for researchers who are interested in studying the factors affecting tourists’ decisions to visit cultural touristic places in Thailand’s southern border provinces. They can build their future research based on the current findings, trying to investigate more factors and policies that promote culture tourism in Southern Thailand in particular and the whole of Thailand in general. They can develop other models to better understand other significant factors of cultural tourists’ behavior towards visiting cultural touristic places. This study broadens the discussion of marketing strategies by showing how approaches like enhancing perceived value, improving physical environments, and fostering tourist engagement can be adapted to other cultural tourism destinations. These strategies should be tailored to fit the cultural and infrastructural context of each location, such as heritage sites in Southeast Asia or Europe. We also integrate sustainable tourism practices, including eco-friendly cultural events, community-led heritage preservation, and low-carbon marketing through social media, aligning with Thailand’s “Triangle of Stability, Prosperity, and Sustainability” plan (SOUTH, 2017). A practical roadmap is provided for stakeholders, recommending actions like forming partnerships with local communities to co-create authentic experiences and investing in sustainable infrastructure. We also link trust-building strategies to UNESCO’s sustainable tourism guidelines, emphasizing community-based engagement as a core component.

Given the significance of promoting cultural tourism, it is essential for entrepreneurs to comprehend the factors influencing tourists’ decisions to visit cultural attractions. This research offers advantages for entrepreneurs in managing activities associated with culture and tradition, including local products, tourism, housing characteristics, tourist attractions, traditional foods, and the lifestyle of villagers, to meet tourists’ expectations. The findings offer a valuable paradigm for understanding the influence of involvement on destination satisfaction and word-of-mouth recommendations, which have substantial implications for destinations’ marketing and development. Increased connections between tourists and the local population, together with chances for tourists to actively participate in destination-related activities, may stimulate their word-of-mouth communications. In addition, other stakeholders, such as the government and policymakers, can also utilize this study to help promote cultural tourism. First, high-quality services and competitive prices boost the perceived value and tourist satisfaction. Second, unique and engaging cultural events that highlight quality of experience can boost satisfaction and WOM. Thirdly, cultural places’ physical environments and infrastructures must be maintained to ensure a positive tourist experience. Fourthly, training residents and service providers to improve their quality of behavior may improve attitudes and satisfaction. Fifth, engaging visitors through interactive and participatory events boosts satisfaction and WOM. Finally, the social impact from online reviews and social media will build trust and attract visitors.

Moreover, cultural site managers may employ effective strategies in the design and development of their enterprises, such as organizing diverse large-scale events and offering distinctive educational experiences. A well-structured presentation of the cultural object, including its location, internal layout, pathways, lighting, and informational panels, may increase visitor interest and engagement. The interpretation and intangible aspects of cultural products can enhance visitors’ comprehension and experiences of cultural tourism. Improving visitors’ attention and engagement can enhance the quality of experience, leading to an increased perceived value and satisfaction among visitors. This inference promotes visitor loyalty. Enhancing the quality of the visitor experience as a management objective and ensuring that this quality translates into perceived value and satisfaction is essential for cultural site managers in the development of sustainability plans. Customer experience has become a significant concept in cultural tourism marketing, as the global experience gained often determines tourist satisfaction. Visitors seeking a comprehensive experience in cultural tourism may pursue leisure, culture, education, and social connections. Cultural site administrators must aim to meet visitors’ expectations regarding engagement, serenity, and educational value, as demonstrated by this study, to provide a holistic quality experience. This conclusion is clear, as many cultural organizations have prioritized public participation in their policies and programming.

8.2. Theoretical Contributions

This research contributes to the culture tourism and hospitality literature by integrating the theories of the S-O-R model, cognitive–affective–conative model, and TPB. This study extends the applicability of the S-O-R framework in integrating the role of the physical environment, perceived value, quality of experience, social influence, and attitudes in predicting WOM intention in the context of cultural tourism. Also, this study incorporates both internal and external aspects of cultural tourists that trigger their WOM intentions. Moreover, this research fills the gap in TPB by improving on its limitations by considering the unbiased determinant of behavior. Hence, this is one of the unique contributions of this study, as TPB was originally based on cognitive processing and ignored one’s needs/motivations before engaging in certain behaviors. In addition, this study also provides various determinants of the WOM intention of cultural tourists. This model was employed to forecast the cultural tourists’ word-of-mouth intentions in a country where limited research has been conducted on the subject. This study also helps to address a big gap in the literature by determining what factors impact their word-of-mouth intentions in this particular environment. This research presents a number of important conclusions about the elements that influence trust, attitude, and satisfaction. PV, tourist engagement, quality of physical environment, and quality of experience are the main findings regarding cultural tourists’ word-of-mouth intentions. It was found that trust was more dependent on social influence. Similarly, perceived behavior control significantly impacts attitude. We compare our findings with studies like Jalilvand et al. (2017) and Rasoolimanesh et al. (2021), noting that our model’s inclusion of tourist engagement and quality of physical environment as predictors of satisfaction adds new dimensions to WOM research. To conclude, the research presented in this study yields significant theoretical advances. This study enhances prior research by establishing a newfound mixed approach that integrates the S-O-R and cognitive–affective–conative models and TPB. Trust, attitude, and satisfaction are the exact determinants of WOM intentions.

9. Limitations and Suggestions for Future Research

This study has limits, notwithstanding its insights. This study mostly addresses Thai cultural tourism, so the findings might not be relevant to other travel or destinations. Future research could investigate these ideas under several conditions to validate the conclusions. Also, this study could be biased by self-reported data. Future research should employ other approaches and observational data to acquire a further understanding. This study employs a cross-sectional design and evaluates intentions at a specific point in time, constrained by temporal limitations. Individuals’ perspectives change as they acquire life experiences. As a result, the longitudinal method is the most effective. The use of purposive sampling, while effective for reaching cultural tourists in this specific region, may limit the generalizability of the results to other types of destinations or tourist segments. We emphasize the unique socio-cultural and geopolitical characteristics of the study area, including its distinct cultural heritage and ongoing stability efforts, and caution against applying these findings to other regions without considering contextual factors. To improve the generalizability, we suggest that future research adopt probability sampling or conduct comparative studies across varied tourism contexts. Future research may shed light on the whole country or extend to the Southern Asian region, too. These findings are most relevant to cultural tourism in heritage-rich, emerging destinations with contexts similar to Thailand’s southern border provinces. We recommend future cross-cultural studies to test the model’s robustness in different settings, such as European heritage sites or urban cultural tourism. This research may establish a foundation for the application of WOM intent within the tourism sector. A full comparison of data, information from other countries, and specifics on the actual application remains insufficient. In the future, scholars may want to investigate this further to elucidate the broader applications of cultural tourism. A more miscellaneous sample of individuals from different educational backgrounds, ages, clusters, and levels of discretionary money will deliver further precise findings. To better understand culture tourists’ interests and the factors that influence their WOM intention, future research should include more social and consumer behavior theories. This study evaluates the scale of word-of-mouth intentions within the framework of cultural tourism, namely in Southern Thailand. A potential avenue for future research is the comparative analysis of tourists to Thailand’s southern provinces vs. those in the northern and central regions, which may yield significant insights into the broader Thai tourism landscape. Additionally, subsequent research must replicate this work in various circumstances. A potential avenue for future research may be comparing tourist participation in heritage tourism with that in other tourism contexts. We have also posited that attitude, trust, and contentment might influence word-of-mouth intentions in a tourism context. Additional research is necessary to evaluate the magnitude of this link and its associations with other behavioral objectives of tourists. Finally, future studies should examine the model qualitatively. Participant interviews or focus groups can reveal more.