Factors Affecting Tourist Satisfaction in Ecotourism: A Case Study of the Phong Nha-Ke Bang National Park, Vietnam

The PNKBNP is a UNESCO World Natural Heritage site, featuring a magnificent cave system and rich biodiversity, making it an attractive destination for ecotourism (Van Mai, Doo-Chul 2020). The national park is located in Quang Binh Province, part of the North Central region of Vietnam. Currently, the national park covers a total area of 85,754 ha, including three main zones: the strictly protected zone, the ecological restoration zone, and the buffer zone (Ly, Xiao 2016). The PNKBNP was established to protect one of the world’s two largest limestone mountain regions, encompassing approximately 300 caves, and to conserve the Northern Truong Son ecosystem in the North Central region of Vietnam. The park is characterised by limestone formations, caves, underground rivers, and rare flora and fauna. The predominant vegetation type is tropical moist evergreen forest on limestone, at an elevation of 800 m above sea level. Forests cover 96.2% of the park’s area, with 92.2% being typical primary forests on limestone mountains, featuring distinctive vegetation types (Khuong, Huong 2014). In recent years, the number of visitors to the PNKBNP has increased rapidly, generating a large amount of data for research (Selcuk et al. 2023). Additionally, compared to other tourist destinations, research on ecotourism at the PNKBNP remains limited. Therefore, to provide further information for researchers, this study selected the PNKBNP as the study area.

Primary data were gathered through a survey administered directly to tourists from June to September 2024. The survey consisted predominantly of closed-ended questions, using a 5-point Likert scale to ensure reliable responses. For instance, tourist satisfaction was assessed on the scale, where 1 indicated ‘strongly unsatisfied’ and 5 represented ‘strongly satisfied’.

After finalising the questionnaire, the next crucial step was to determine the sample size to ensure the survey results were highly reliable. Methods such as regression and exploratory factor analysis typically require a large sample size to ensure high research reliability. However, owing to time and financial constraints in collecting a substantial amount of data, this study determined the sample size based on factor analysis methodology. The observation-to-measurement ratio is 5:1, meaning that each measure requires at least five observations (Hair et al. 1998). The required number of observations was determined using the following formula: N=5×mN = 5 \times m where: –

N: number of observations;

Ultimately, the survey yielded 185 valid questionnaires. The collected data were subsequently digitised, organised, tabulated, and analysed using the SPSS 26.0 software (IBM Corporation).

Cronbach’s alpha method measures unsuitable variables and reduces noise variables in the research process by evaluating the scale using Cronbach’s alpha reliability coefficient. It measures the extent to which items on a scale measure the same concept (Agbo 2010). The higher the Cronbach’s alpha coefficient, the greater the reliability of the scale. Regarding the reliability of the observed variables, they were considered reliable when the corrected item-total correlation coefficient was ≥0.3 (Bland, Altman 1997). Variables with an item-total correlation coefficient of <0.3 will be removed. The scale with a Cronbach’s alpha coefficient of 0.6 or higher can be used in cases where the concept being studied is new (Tavakol, Dennick 2011). Regarding the reliability of the measurement scale, Cronbach’s alpha of 0.7 to nearly 0.8 indicates an acceptable measurement scale, while Cronbach’s alpha of 0.8 to almost 1 indicates a good measurement scale (Cortina 1993).

The exploratory factor analysis is a statistical technique used to identify hidden structures (also known as factors) from a large set of observed variables (Cudeck 2000). The primary objective of the exploratory factor analysis in this study is to simplify the data collected from multiple survey questions, uncover key latent factors influencing tourist satisfaction, and evaluate the significance of each factor. This approach provides a deeper understanding of tourists’ needs and enables effective management decisions to enhance service quality (Correia et al. 2008).

To assess whether the dataset is suitable for factor analysis, the following criteria are applied: –

Kaiser–Meyer–Olkin (KMO) measure: The index measures the adequacy of data for factor analysis. The value between 0.5 and 1 is considered appropriate for factor analysis (Kaiser, Rice 1974).

Multivariate regression analysis is a widely used statistical method for modelling the relationship between a dependent variable and one or more independent variables (Harrell 2001). It helps understand how independent variables affect the dependent variable (Chadee, Mattsson 1996, Suanmali 2014). By identifying key factors, regression analysis enables a deeper understanding of what tourists expect and value. This allows service providers in the tourism industry to develop forecasting models, make informed business decisions, and improve service quality effectively.

To measure and evaluate the influence of the factors on tourist satisfaction, a multivariate regression method is used. The multiple regression model is represented by Eq. (1): 1Tourist satisfaction =β0+β1×F1+β2×F2+…+βn×Fn+ε{\rm{Tourist satisfaction }} = {\beta _0} + {\beta _1} \times {F_1} + {\beta _2} \times {F_2} + \ldots + {\beta _n} \times {F_n} + \varepsilon where: –

β₀: regression constant;

Demographic information obtained from the survey sample is presented in Table 2. Domestic tourists accounted for 53.0%, indicating the significance of the domestic tourism market. However, the number of international tourists was also substantial, suggesting the potential for international tourism development. In terms of gender, 58.3% of tourists were female, while the remaining 41.7% were male. Regarding age, the majority of respondents were between 29 and 48 years old, implying a higher capacity to afford ecotourism; followed by the age group of 49–65. Those with a college degree constituted 61.9% of the sample, suggesting an association between ecotourism and higher educational attainment, followed by 21.85% who indicated having only a high school degree. Regarding occupation, the majority were private employees, indicating a higher capacity to afford tourism, followed by students, reflecting the younger generation’s inclination towards nature exploration.

Cronbach’s alpha was used to examine the reliability of factors affecting tourist satisfaction. In other words, this technique is employed to determine whether the observed variables align with the underlying concept of a factor. After conducting Cronbach’s alpha, four observed variables were excluded (EG1, EG2, EG3, and PF1) due to item-total correlation coefficients of <0.3.

The second test confirmed the reliability of the scales, as Cronbach’s alpha coefficients were relatively high (>0.7). Additionally, all 30 remaining observed variables had total variable correlation coefficients greater than 0.3, further supporting their reliability. Cronbach’s overall alpha value was 0.806, indicating that 80.6% of the variance in the combined 30 items represents true score variance, demonstrating strong internal consistency.

The study continued to analyse Cronbach’s alpha coefficients for each independent variable. However, the independent variable ‘entertainment and shopping’ was excluded from the model because its observed variables had been removed in previous steps. All remaining variables had a Cronbach’s alpha value greater than 0.7, indicating that the observed variables within each variable consistently measured a common concept and thus can proceed with further evaluation (Table 3).

The first step in applying exploratory factor analysis is to evaluate whether the dataset meets the necessary conditions. The results show that the KMO measure was 0.8, indicating that the variables in the dataset were sufficiently correlated to conduct factor analysis. Additionally, Bartlett’s test was statistically significant (p-value <0.05), confirming that the correlation matrix of the variables significantly differed from the identity matrix, meaning that the variables were interrelated.

Using the principal component analysis extraction method with Varimax rotation for independent variables, five factors were extracted from the 30 observed variables at an eigenvalue of 1.069 (representing five latent factors affecting tourist satisfaction). Furthermore, the TVE showed that these five factors accounted for 70.4% of the dataset’s variance, exceeding the minimum threshold of 50%, thus confirming the model’s adequacy. Finally, in the rotated component matrix, all factors had factor loadings greater than 0.5, meeting the requirements. As a result, no factors needed to be removed from the scale (Table 4).

After identifying the factors, the next step is to assign names and interpret their meanings. This process involves analysing the observed variables with high factor loadings on the same factor, as these variables define the underlying concept of each factor. Based on the characteristics of the grouped observed variables from independent variables, a new name for the factor is determined (this attribute is known as the exploratory attribute of factor analysis) (Table 5). –

Factor 1: Renamed as ‘destination attraction’; this factor includes observed variables SS1, SS2, SS3, TS3, and FB3.

Based on the factor analysis results, the research model has been adjusted to include five factors: (1) destination attraction, (2) ecotourism services, (3) staff quality, (4) price, and (5) infrastructure. Tourist satisfaction remains the dependent variable, while the newly identified components serve as the independent variables in the revised model. Accordingly, the proposed hypotheses are presented in Table 6.

The proposed hypotheses provide a more comprehensive understanding of the factors influencing tourist satisfaction. By focusing on these specific factors, the research can delve deeper into the individual effects of each factor on overall tourist satisfaction at the PNKBNP.

The results showed an adjusted R² value of 0.840. This indicates that the independent variables in the model explain 84.0% of the variation in the dependent variable (Table 7). The remaining 16% is attributed to other factors not included in the model.

Additionally, to assess the overall fit of the regression model, the study examines the F-value using the analysis of variance (ANOVA) (Table 8). The F-value is 53.082, and the Sig-value is 0.00, indicating that the multiple regression model is suitable for the dataset and can be used.

The variance inflation factor (VIF) for each factor is less than 10 (Table 9), indicating that the regression model does not violate the multicollinearity phenomenon (independent variables are highly correlated).

After the conditions for multiple regression analysis were met, the results determined that there is a significant linear relationship between the five factors and tourist satisfaction (p-value <0.05). Among the five factors, ‘destination attraction’ had the strongest impact on tourist satisfaction at the PNKBNP (β = 0.322), followed by ‘ecotourism services’ (β = 0.213), ‘infrastructure’ (β = 0.096), and ‘price’ (β = 0.076). Although ‘staff quality’ showed the weakest effect (β = 0.009), all corresponding hypotheses (H1–H5) were statistically supported (Table 10).

Based on the identified regression coefficients, Eq. (1) can therefore be re-expressed as Eq. (2): 2Tourist satisfaction = 0.197 + 0.322 × destinationattraction + 0.213 × ecotourism services + 0.096 × infrastructure + 0.076 × price + 0.009 × staffquality{\rm{Tourist satisfaction = 0}}{\rm{.197 + 0}}{\rm{.322 \times destination}}\;{\rm{attraction + 0}}{\rm{.213 \times ecotourism services + 0}}{\rm{.096 \times infrastructure + 0}}{\rm{.076 \times price + 0}}{\rm{.009 \times staff}}\;{\rm{quality}}

Therefore, the final research model of factors affecting tourist satisfaction at the PNKBNP has been established (Fig. 2). The model not only highlights the relative influence of each factor but also serves as a useful framework for policymakers, tourism managers, and stakeholders in designing appropriate strategies to enhance tourist experiences and promote sustainable ecotourism development in the PNKBNP.

Fig. 2.

The final model of the study.

The attractiveness of a destination is a fundamental factor that draws the interest of tourists and tourism managers (Jiménez-García et al. 2020). ‘Destination attraction’ has the strongest impact on the satisfaction of ecotourists at the PNKBNP because it is a destination with outstanding natural value and scenic beauty, meeting tourists’ needs for pristine natural experiences. This underscores that the core value of ecotourism lies in the beauty and uniqueness of nature, which is the primary factor contributing to the highest levels of tourist satisfaction. The PNKBNP is renowned for its magnificent cave systems, karst landscapes, pristine forests, and diverse ecosystems (Ly, Xiao 2016). These natural features are exceptionally unique and difficult to find elsewhere. Ecotourists often seek close-to-nature experiences, and the pristine, captivating beauty of this destination is the main reason they choose it. Beyond natural landscapes, the PNKBNP also boasts significant cultural, historical, and UNESCO World Heritage values (Tran 2020). Tourists come not only to admire the natural beauty but also to gain insights into local history and culture. The combination of natural and cultural factors enhances the attractiveness of the destination, thereby increasing overall satisfaction.

Previous research indicates that service quality and satisfaction are distinct but closely related concepts, with customer satisfaction seen as an outcome and service quality as the reason; satisfaction is predictable and service quality is an ideal standard (Hau 2014). Oliver (2014) suggests that service quality should influence customer satisfaction. The better the service meets customer expectations, the higher the satisfaction. According to Parasuraman et al. (1994), service quality is determined by multiple factors and is a key determinant of customer satisfaction. If a service provider delivers high-quality products that meet customers’ needs, the business will initially ensure customer satisfaction. Spreng and Mackoy (1996) also assert that service quality can directly impact the overall experience of tourists. Good service increases tourist satisfaction, whereas poor service may diminish it.

Although the primary goal of ecotourists is to explore and experience nature, basic tourism services such as transport, accommodation, and dining remain crucial. The PNKBNP is a pristine area with challenging terrain, making transport services particularly important. The quality of transport services, especially flexibility and safety, significantly affects the tourist experience by enabling access to hard-to-reach areas like caves or primary forests without encountering unforeseen risks or difficulties. Moreover, ecotourism activities in the PNKBNP often require physical exertion, such as climbing, cave exploration, or kayaking. After these activities, tourists need comfortable accommodation and quality dining services to recuperate. Furthermore, while ecotourists may accept some limitations in amenities compared to other types of tourism, they still require basic services to feel satisfied. Comfortable lodging and safe dining services that meet hygiene and quality standards contribute to a better experience and enhance satisfaction after days of nature exploration.

Numerous studies have addressed the close relationship between infrastructure and tourist satisfaction (Sugiama et al. 2022). The infrastructure component of tourism development is crucial as it supports the competitive advantage of a destination. Furthermore, the development of comprehensive public infrastructure is essential for high-quality tourism facilities at tourist destinations (Jusoh et al. 2013). This research indicates that while infrastructure has a significant impact, it is not the primary factor influencing tourist satisfaction at the PNKBNP. This aligns with the characteristics of an ecotourism destination, where natural attractiveness and nature-based experiences play a more prominent role (Donohoe 2011, Cobbinah 2015, Stronza et al. 2022). Tourists visiting a nature reserve may accept less comfortable infrastructure compared to urban or resort destinations. They may be willing to overlook infrastructural shortcomings if natural tourism experiences offer substantial value. Thus, while infrastructure is important, it is not the foremost factor. Additionally, tourism infrastructure depends heavily on the development of other factors, such as tour guiding activities, tour organisation processes, and access to key sightseeing areas. If these factors are optimised, tourists will experience greater satisfaction with the overall trip, reducing the reliance on the quality of infrastructure.

Price is always a sensitive factor for tourists (Oliver 2014, Carlos Castro et al. 2017). Tourist satisfaction depends not only on the amount spent but also on the balance between costs and the value received from the experience (Giao et al. 2021). If tourists feel that what they receive – ranging from natural landscapes to service quality – is commensurate with the amount paid, they will feel satisfied. However, because the focus of ecotourism is on experiences, the price factor does not play as significant a role as other factors. Furthermore, the PNKBNP attracts a diverse range of tourists, from international visitors to domestic travellers, including those who are willing to pay high prices for quality services and those with limited budgets (Van Mai, Doo-Chul 2020). This results in varying levels of importance of the price factor for different groups. For those seeking unique travel experiences, price may not be the most critical factor influencing overall satisfaction. Additionally, tourists have a variety of service options, from luxury accommodations to more budget-friendly services, accommodating different budget levels. This diversity helps tourists easily find options that fit their financial capabilities, thereby reducing the impact of price on overall satisfaction.

This study indicates that staff quality has a minimal impact on overall tourist satisfaction at the PNKBNP. This can be explained by the fact that ecotourists at the PNKBNP are generally more concerned with factors directly related to nature experiences and tourism services rather than staff quality. Additionally, many activities in ecotourism at the PNKBNP are self-guided experiences, where tourists participate in activities such as trekking, cave exploration, or primary forest hikes on their own (Van Mai, Doo-Chul 2020, Selcuk et al. 2023). The role of staff in these activities is primarily supportive, such as providing information or basic guidance. Therefore, even if the quality of staff is not high, tourists can still explore and experience on their own, making staff quality a non-essential factor in determining satisfaction. Despite its minimal impact, staff quality still plays an important supportive role in ensuring a smooth and comfortable experience for tourists (Han, Hyun 2015, Giao et al. 2021). Tasks such as guiding, service, and safety assurance contribute to maintaining a seamless experience. If the staff quality factor is neglected or not given adequate attention, tourists may encounter difficulties during their visit, which could negatively affect overall satisfaction.