Bintan Regency is a regency in the Riau Islands Province, located close to Singapore and Malaysia. This makes Bintan a popular tourist destination for international tourists, especially those from Singapore and Malaysia. Top tourist attractions in Bintan Regency include Lagoi, Sakerah Beach, and Trikora Beach. It also boasts stunning underwater biodiversity and a wealth of coastal resources with potential as marine tourism attractions, such as mangrove ecosystems, coral reefs, and turtle nesting sites (Herawati et al., 2023). Bintan also boasts a rich history and culture, one of which is Penyengat Island, a historic island that was once the center of the Johor Riau Lingga Kingdom. Penyengat Island is also being developed as a global center for Malay-Islamic studies. Bintan regularly hosts international events to attract tourists, including the Bintan Triathlon, Tour de Bintan, Bintan Marathon, and others.

Despite its rich natural resources, historical sites, diverse cultural heritage, and a calendar of international events such as the Bintan Triathlon, Bintan Regency continues to face challenges in its tourism sector. Data from the Bintan Culture and Tourism Office shows significant fluctuations in international tourist arrivals. For example, while international tourist arrivals reached 573,579 in 2019, they dropped dramatically to 64,896 in 2020 due to the pandemic, showing a slow recovery to 222,101 in 2023 and 208,590 in 2024. This volatility has a direct impact on local government revenue from tourism.

Furthermore, hotel room occupancy in Bintan remain consistently low, averaging 52.63% in 2024, well below ideal industry benchmarks. The average length of stay for international tourists in Bintan is also very short, at only 1.92 days for star-rated hotels and 1.65 days for non-star-rated hotels in 2024, as seen in Table 2. This indicates that tourists are not yet fully engaged with the destination's offerings. According to the UNWTO, the average length of stay for international tourists globally is 7 to 10 days for resort and beach destinations, 5 to 8 days in Asia Pacific, and 2 to 4 days for travelers traveling within a geographically close or short distance. Bintan's length of stay figures, which are well below the global average, even for short-haul travelers, indicate that Bintan Regency is failing to attract longer-stay tourists.

This problem is further exacerbated by specific perceived risks. For example, high Visa on Arrival (VoA) fees and oil spill pollution incidents. This is demonstrated by the consistent number of participants in the Bintan Triathlon from 2017 to 2019, reaching 1,300. The event was postponed due to the pandemic and was re-organized in 2023, reaching the target of 500 participants. However, the number of participants declined again in 2024. A statement from the General Manager of PT Bintan Resort Cakrawala confirmed that the VoA fee, which burdens international tourists visiting Bintan in the short term, was the primary cause of the decline (Antaranews, 2023). This significantly contributes to negative perceptions that can deter potential tourists. To address these challenges, it is crucial for Bintan to maximize its tourism potential, as seen in Table 1.

Table 1. Number of Bintan Triathlon Participants 2017 - 2024

Sumber: Data Diolah (2025)

Research exploring destination attractiveness and travel motivations conducted by (Maghrifani et al., 2021; Yacob et al., 2019) has focused primarily on developed, mature destinations with robust infrastructure, such as Bali or Phuket. In contrast, Bintan, as an emerging island destination, is characterized by unique development stages and challenges that may not be adequately captured by existing models. Furthermore, previous research (Charag et al., 2020; Jarumaneerat, 2022) predominantly positions perceived risk as an independent or mediating variable.

This study builds on this by examining the crucial role of perceived risk as a moderator, hypothesizing that risk influences the strength and direction of the relationships between destination attractiveness, travel motivation, and visit intention. This innovative theoretical approach, coupled with its focus on an under-researched, emerging destination, addresses a significant gap in the current tourism literature. The identified challenges underscore the urgent need for a deeper understanding of the factors influencing international tourists' travel intentions in Bintan. By investigating the direct influence of destination attractiveness and travel motivation, and most importantly, the moderating role of risk perception, this study aims to provide new theoretical insights into destination marketing in a developing archipelago context. Practically, these findings will provide strategic guidance for Bintan tourism stakeholders— including local governments, tourism operators, and destination management organizations—to develop more effective and risk-aware marketing strategies. This research is expected to contribute to mitigating existing fluctuations in tourist arrivals and promoting sustainable tourism development in Bintan Regency.

Push and Pull Motivation Theory

The Push-Pull framework explains how internal drivers and destination- specific attributes shape tourism (Crompton, 1979). These factors are intrinsic factors that motivate individuals to escape from daily routines and seek new experiences. These factors include escaping routine habits or stress, exploring, social interaction, self-discovery through self-development, social status, rest through relaxation, cultural and educational experiences, spiritual goals, and health and well-being.

Tourists' decisions to visit a particular destination are influenced by external attractions, which are uncontrolled by the tourist, but are influenced by the destination's presence, such as natural attractions, culture, festivals, climate conditions, recreational facilities and activities, accessibility, affordability, hospitality, service quality, government policies, tourism promotions, shopping, and entertainment. The Push–Pull framework clarifies how internal drivers and destination-specific attributes shape tourism decisions

An individual's perception or assessment of a destination and its ability to meet their travel or vacation needs is often referred to as destination attractiveness (Hu and Ritchie, 1993) in Dey et al. (2020). Kresic and Prebezac (2011) in Charag et al. (2020) argue that destination attractiveness is a mental image of a destination formed based on its physical appeal. This study adopted the components of destination attractiveness including natural attractions, historical and cultural attractions, rural attractions, sports and recreation opportunities, destination location and transportation, accommodation and catering, and social conditions (Dey et al., 2020). Ouariti and Jebrane (2020) found that the more tourists believe a destination can meet their needs and desires, the more attractive the destination is to visit. Chaudhary and Islam (2020) found that destination attractiveness has a positive effect on tourists' satisfaction and intention to visit in the future. (2020) found that several attributes of tourist destination attractiveness influence tourists' choice of homestay.

Travel Motivation

Many theories and models explain motivation in the context of tourism, including the push and pull theory by Crompton (1979), the tourist typology model by Plog (1974), the escape model by Dunn Ross and Iso-Ahola (1991), and many others. This study examines tourist travel motivation in relation to other variables such as Perceived Risk and travel intention.

Crompton (1979) suggests that motivation is a driving factor that drives people to travel. Chi and Phuong (2022) found that travel motivation, time perspective, and destination image influence travel intention. Maghrifani (2021) found that travel motivation positively influences travel intentions, mediated by destination image. These findings also identified four dimensions of travel motivation: novelty, escape, certainty, and interaction.

Perceived Risk

2Perceived risk was introduced by Bauer (1960), who defined it as a consumer's perception of uncertainty or potential loss resulting from the purchase of goods or services. Jacoby and Kaplan (1972) developed this theory by dividing perceived risk into several categories: financial, performance, physical, psychological, and social. Jacoby and Kaplan argued that these components are independent, meaning that if one risk increases, other risks may also increase, decrease, or even remain unaffected.

Specifically in the context of tourism, Han et al. (2005) proposed seven dimensions of travel risk: value, health, terrorism, equipment, communication, social, and psychological. This study adopts the risk aspects from Jarumaneerat (2022), which combine several aspects of travel risk from previous studies: physical, health, financial, performance, service, communication, socio- psychological, time, natural disasters, political uncertainty at the destination, and satisfaction of expectations.

Visit Intention

According to Su et al. (2020), travel intention is the subjective tendency of potential tourists to visit a destination, triggered by internal and external information. Chaudhary & Islam (2020) found that the attractiveness of a tourist destination positively influences tourists' satisfaction and future travel intentions. Furthermore, tourists' intrinsic motivation can influence their interest in visiting a destination (Chi & Phuong, 2022; Maghrifani et al., 2021). However, Falahuddin et al. (2020), referring to Maslow's theory, argue that there is a relationship between travel intention and perceived risk, namely physiological, safety, social, higher-order needs, and self-actualization. The desire to fulfill needs is accompanied by the risk of not fulfilling those desired needs.

Pranindyasari et al. (2023) argue that perceived risk makes people consider avoiding travel to a tourist destination. Carvalho's (2022) findings indicate that perceived risk is a determinant of future travel intentions. Caber et al. (2020) found that perceived risk negatively moderated the relationship between travel motivation, destination image, and visit intention in Spain and Greece. Meanwhile, Manchanda and Deb (2022) found that perceived risk was a variable that negatively moderated the effect of experiential marketing on future visit intention. However, Lin et al. (2012) in Caber et al. (2020) showed different results, indicating that respondents in this study were more likely motivated by risk factors when visiting Australia.

The object of this research is tourist behavior, specifically international tourists. This study tests and analyzes a SEM structural model with independent variables of destination attractiveness and travel motivation, moderating variables of risk perception, and dependent variables of visit intention. The research methods used are shown below.

Table 2. Research Methods

Source: Author’s Compilation

Operational and Measurement Variable Definitions

The operational definition and measurement of each variable used a 5- point Likert scale. Destination attractiveness is a destination's appeal to tourists, encompassing dimensions such as nature, history and culture, countryside, sports and recreation opportunities, transportation, accommodation, and social amenities (Dey et al., 2020).

Travel motivations are the factors that motivate a person to visit a destination (Dann, 1977 in Maghrifani et al., 2021), encompassing the dimensions of seeking interaction, novelty, escape, and assurance. Risk perception is an individual's subjective assessment of the potential for unpleasant events during a trip (Jarumaneerat, 2022), encompassing physical, health, financial, service and performance, communication, socio-psychological, time, natural disasters, political unrest, and expectation-satisfaction risks.

The operational definition of the visiting intention refers to the desire to visit a destination (Zhao et al., 2024), encompassing the dimensions of near- term visit intention, information-seeking intention, future visit intenstion.

SEM-PLS version 3.0 was chosen due to its ability to handle complex models with reflective and formative constructs, its suitability for non-normal data, and its effectiveness with smaller sample sizes compared to covariance- based SEM. This analysis involves a two-step approach (1) evaluation of the outer model to assess reliability and convergent and discriminant validity, (2) evaluation of the inner model to assess the significance of path coefficients, R- squared values, and predictive relevance (Q-squared). Moderation analysis uses the SmartPLS interaction approach, by multiplying the independent variables (destination attractiveness, travel motivation) by the moderating variable of risk perception.

Respondent Characteristics

Based on data from 200 questionnaires, 116 respondents, or 58% of the total, were male. The majority of respondents (38 or 19%) came from Singapore, followed by 37 or 18.5% from Malaysia, and 22 or 11% from China. Other respondents came from India, South Korea, Japan, the Philippines, Australia, the United Kingdom, the United States, Thailand, Bangladesh, Taiwan, the United Arab Emirates, Vietnam, and Russia. 50 or 25% were students, 32 or 16% were entrepreneurs, and 28 or 14% were private sector employees. Other occupations of respondents included freelancers, IT workers, teachers, engineers, retirees, doctors, housewives, hospitality workers, photographers, government employees, content creators, and nurses. The majority of respondents had a high annual income, with 119 respondents, or 59% of the total. Five percent of the respondents had an annual income above $14,005, followed by middle-income respondents (53 respondents, or 26.5% of the total), and low-income respondents (28 respondents, or 14%). This classification is based on the Gross National Income (GNI) per capita figures released by the World Bank, which defines incomes below $1,145 as low-income, incomes between $1,145 and $14,005 as middle-income, and incomes above $14,005 as high-income.

Outer Model

Reliability and validity values for measuring the model. Construct loadings exceeding the 0.7 threshold indicate good convergent validity. Average variance extracted (AVE) values for constructs above 0.5 further support convergent validity. Discriminant validity is confirmed because the square root of the AVE of each construct is greater than its correlation with other constructs (Fornell-Larcker Criterion), with the HTMT ratio below the 0.90 threshold. Composite reliability and Cronbach's alpha values for all constructs exceed 0.7, indicating adequate internal consistency reliability, as shown in Table 3.

Table 3. Reliability Test Results

Source: Data Processing Results (2025)

Inner Model

The structural model explained a substantial proportion of the variance in visiting intention, with an R-squared value of 0.853, indicating that destination attractiveness, travel motivation, and risk perception collectively explained 55% of the variance in international tourists' visit intentions. The Q- squared value of 0.824 indicated good predictive relevance of the model. The results of the analysis to complete the structural model between variables can be seen below.

Table 4. The Relationships between Variables

Hypothesis testing is performed by examining the relationship in PLS- based SEM using bootstrapping simulation and assessing the significance level of the structural model using the T-Statistic and P-Value. If the T-Statistic value is greater than 1.96 and the P-Value is less than 0.05, the hypothesis is accepted. The structural model appears as below.

Figure 1. Struktural Equation Model of the Study

The Effect of Destination Attractiveness on Visiting Intention (H1)

The analysis results show that tourists' underlying destination attractiveness substantially influences their likelihood of visiting. This is evidenced by the significant positive influence of destination attractiveness on international tourists' visit intention. The path coefficient is 0.426, with a t-statistic of 13.547 and a p-value of 0.000. Since the p-value is less than 0.05 (or t-statistic > 1.96), Hypothesis 1 is supported. This finding indicates that as Bintan’s perceived attractiveness increase, so does international tourist’s intention to visiting,

These results align with existing literature emphasizing pull factors in tourism motivation, where attractive destination attributes are a key driver of tourist choice (e.g., Dey et al., 2020). This confirms the relevance of Destination Attractiveness as a fundamental construct in influencing visiting intention, even in developing island destinations like Bintan. This reinforces the view of Goeldner et al. (2000) stated that destination attractiveness is a core component of the tourism system, requiring the strengthening of natural, cultural, and social aspects to create a holistic experience.

For Bintan's tourism stakeholders, these findings underscore the importance of continued efforts to enhance and promote its unique tourist attractions. Focus should be placed on Bintan's natural beauty (e.g., beaches, mangrove ecosystems), cultural heritage (e.g., Penyengat Island), and, in particular, sports and recreational opportunities (e.g., golf, water sports), as these have been identified as key aspects of Bintan's attractiveness. Particular attention should be paid to ensuring that advertised recreational facilities are fully operational and well-maintained to avoid negative perceptions. Furthermore, improving accessibility and providing multilingual information could significantly enhance this attractiveness.

The Effect of Travel Motivation on Visiting Intention (H2)

The analysis results indicate that tourists' underlying travel motivation substantially influences their likelihood of visiting. This is evidenced by the significant positive effect between travel motivation and international tourists' visit intention. The path coefficient is 0.354, with a t-statistic of 8.026 and a p- value of 0.000. Hypothesis 2 is supported. This indicates that a stronger motivation to travel positively influences the intention to visit Bintan.

This finding aligns with Crompton's (1979) Push and Pull Theory of Motivation, which states that internal desires (push factors) significantly drive an individual's decision to travel. This confirms that intrinsic motivations such as seeking novelty, escape, certainty, or interaction are strong precursors to visit intention in the tourism context. This supports the notion that understanding these underlying motivations is crucial for predicting tourist behavior.

To capitalize on this, Bintan's tourism sector must develop targeted marketing campaigns and tailor tour packages that cater to the diverse motivations of international tourists. For example, designing packages for relaxation seekers (e.g., spa and yoga retreats), adventure enthusiasts (e.g., diving, kayaking, trekking), or cultural explorers (e.g., tours to Penyengat Island, local workshops) can effectively appeal to specific segments. Utilizing emotional storytelling in promotions that highlight personal experiences can further increase engagement and conversions. Addressing short stays (an average of 1.92 days for star-rated hotels) by designing these packages for 3-4 days can also encourage longer visits, in line with UNWTO standards for short-haul travelers.

The Moderating Role of Perceived Risk on Destination Attractiveness and Visiting Intention (H3)

The analysis results indicate that the moderating effect of perceived risk on the relationship between destination attractiveness and tourists' visit intention is statistically proven. The interaction path coefficient is 0.081, with a t-statistic of 3.201 and a p-value of 0.001. Since the p-value is less than 0.05, Hypothesis 3 is supported. This indicates that perceived risk significantly strengthens the positive influence of destination attractiveness on visit intention in this context.

This finding offers a nuanced contribution to the literature on perceived risk, going beyond its common depiction as a purely negative deterrent. It suggests that for certain segments of travelers, particularly allocentric travelers who constitute the majority of respondents (e.g., college students), the presence of perceived risk, when combined with strong destination attractiveness, can be viewed as an element of adventure or novelty, thus strengthening visit intention. This aligns with studies showing that some travelers are driven by a desire for challenging experiences (e.g., Lin et al., 2012, cited in Caber et al., 2020). This highlights a complex interaction where perceived benefits (attractiveness) outweigh perceived costs (risk) in the decision-making process.

While this moderation represents a unique dynamic for some segments, it does not negate the need for risk mitigation for the broader market. For the allocentric segment, Bintan can strategically promote certain "adventure risks" in a controlled and safe manner, such as trekking through mangrove forests or exploring remote islands. However, for the majority of international tourists, addressing perceived core risks remains crucial. This includes advocating for reduced Visa on Arrival (VoA) fees, which are perceived as a financial risk, conducting regular beach cleanups to combat oil spill pollution (a physical risk), and increasing the availability of clear, multilingual information at tourist centers and attractions. These actions are crucial to ensuring Bintan's appeal is not eroded by avoidable negative perceptions, ultimately leading to greater tourist attraction.

The Moderating Role of Perceived Risk on Travel Motivation and Visit Intention (H4)

The analysis results indicate that the role of perceived risk as a moderator between travel motivation and visit intention was not statistically validated in this study. The path coefficient for this interaction was 0.033, with a t-statistic of 1.288 and a p-value of 0.200. Because the p-value was greater than 0.05 (or t- statistic <1.96), Hypothesis 4 was not supported. This indicates that tourists'' motivational strength to travel to Bintan is neither significantly strengthened nor reduced by their risk perceptions.

This finding contributes to the literature by demonstrating that while perceived risk can influence perceptions of destination attributes, it may have a less pronounced impact on underlying intrinsic travel motivations. This suggests that if tourists internal desire to travel is strong (i.e., a strong motivating factor), they may be less susceptible to the modulating effects of perceived risk compared to how it influences evaluations of destination characteristics. This study offers a nuanced view of the interaction between motivators and perceived risk in the context of travel intentions.

The non-significant moderation suggests that strong travel motivation alone may not be sufficient to drive travel if perceived risks are high or unaddressed, as such motivation is not reinforced by risk mitigation efforts. Therefore, focusing solely on appealing to motivation through marketing may not yield optimal results if underlying risk issues (e.g., safety, hygiene, visa issues) are not well managed. Tourism stakeholders in Bintan should ensure that while promoting experiences that align with tourists' motivations, they simultaneously strive to mitigate and clearly communicate perceived risk mitigation efforts to convert intention to actual visitation, especially for risk- averse tourists.

Overall theoretical implications: This contribution enriches the tourism literature linking risk perception to a direct mediator role. It demonstrates its moderating effect on destination attractiveness, particularly for certain traveler segments (e.g., allocentric travelers) who may perceive some risk as part of the adventure. It also contributes to a more nuanced perspective on traveler decision- making. This research validates the push-pull theory for the emerging Bintan region. Furthermore, it highlights the simultaneous importance of intrinsic motivation and extrinsic destination attributes in shaping travel intentions. Furthermore, the finding that risk perception does not moderate travel motivation provides a clearer distinction between how different aspects of traveler psychology are influenced by risk perception, thus enriching theoretical models of tourist behavior.

Overall, these findings offer practical and actionable insights for Bintan tourism stakeholders to increase international tourist arrivals and foster sustainable growth. It is clear that while enhancing Bintan's natural and cultural attractions and accommodating diverse travel motivations are crucial, effective management and communication of perceived risk are also crucial. Prioritizing reducing barriers such as high Visa on Arrival (VoA) fees, ensuring transparent access to information, maintaining cleanliness (e.g., addressing oil spills), and improving safety perceptions are vital steps to unlocking Bintan''s full tourism potential. Tailoring promotional strategies to specific traveler segments (e.g., emphasizing adventure for destination-oriented travelers while ensuring safety for others) will be key to converting interest into actual visits and increasing hotel occupancy rates and low lengths of stay.

The results of the investigation of the influence of destination attractiveness and travel motivation on the interest of foreign tourists in Bintan Regency, with risk perception acting as a moderator, the findings confirmed that destination attractiveness (β = 0.426) and travel motivation (β = 0.354) significantly and positively influence visit intention. Most importantly, risk perception significantly moderates the relationship between destination attractiveness and visit intention (β = 0.081), indicating that for certain tourist segments, risk can unexpectedly increase attraction. However, risk perception was not found to moderate the influence of travel motivation (β = 0.033). This model explains the R-squared value of 0.853% of the variance in visit intention, with good predictive relevance (Q2 = 0.824).

Theoretically, this research advances the understanding of perceived risk by empirically demonstrating its nuanced moderating role, particularly how it can interact with destination attractiveness to influence visiting intention in emerging island contexts. This extends existing push-pull theories by highlighting the complex interplay between extrinsic destination attributes, intrinsic motivations, and risk perceptions.

Managerially, these findings provide actionable insights for Bintan's tourism stakeholders. To boost foreign tourist arrivals and address fluctuating visits, low occupancy, and short stays, strategies must focus on enhancing and promoting Bintan's core attractions while proactively managing perceived risks. Specifically, efforts to reduce financial risks (e.g., advocating for lower Visa on Arrival fees) and mitigate physical risks (e.g., maintaining cleanliness from oil pollution, ensuring safety) are paramount. Simultaneously, tailoring promotional content that aligns with diverse tourist motivations and, for specific segments, strategically framing certain 'adventure risks' can optimize marketing efforts.

To broaden the understanding of factors influencing visiting intention in Bintan, future research can explore several aspects. First, longitudinal studies are needed to monitor changes in visitation intention, destination attractiveness, travel motivation, and perceived risk over time, considering seasonal and global event impacts. Second, comparative studies with destinations like Phuket or Bali can identify competitive advantages and areas for improvement. Third, research should focus on specific tourist segments (e.g., families, solo travelers, or regional markets) to tailor strategies effectively. Fourth, exploring the relationship between sustainable tourism practices and tourist satisfaction is relevant, especially given environmental concerns like oil spills.

Additionally, future studies could incorporate variables such as tourist satisfaction to test its mediating role in the existing relationships, as satisfaction directly impacts length of stay (currently 1.92 days). Analyzing revisiting intention as a dependent variable is also crucial to understand loyalty drivers, vital for increasing hotel occupancy rates (currently 27.18%) and supporting Bintan's tourism recovery. The inclusion of these variables would further expand Push and Pull Motivation Theory by exploring traveler experience and loyalty dynamics in island destinations. Challenges in data access may arise, suggesting collaboration with tourism institutions.

The author conveys deep appreciation to the corresponding author for their steadfast guidance and expertise throughout this research. Sincere thanks are extended to the academic supervisors at Bali Tourism Polytechnic for their insightful advice and encouragement, as well as to the institution itself for providing invaluable opportunities to conduct this study. Special gratitude goes to Mr. Ucu for generously serving as an interview resource person. Lastly, the author is profoundly grateful to all individuals who contributed to the success of this work, whose support, though too numerous to name individually, was instrumental in its completion.