Service quality and environmental sustainability in the vietnamese hotel industry impacts on customer satisfaction and loyalty

VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY _______***_______ DANG THI XUAN SERVICE QUALITY AND ENVIRONMENTAL SUSTAINABILITY IN THE VIETNAMESE HOTEL INDUSTRY: IMPACTS

INTRODUCTION

Research rationale

The hospitality industry represents a fundamental pillar of the global economy, contributing notably to both GDP and employment worldwide As of 2019, the industry supported approximately 334 million jobs and accounted for 10.4% of global GDP (WTTC, 2020) Despite its economic importance, the sector poses significant environmental challenges According to the United Nations World Tourism Organization (UNWTO, 2011), the hotel sector alone contributes around 1% of global carbon emissions, largely due to excessive energy use, high water consumption, and substantial waste generation In light of growing climate concerns, the environmental burden imposed by this industry underscores the urgency of adopting transformative and sustainable approaches

Over the past few years, the hospitality sector has witnessed a rising emphasis on environmental sustainability, driven by increased consumer awareness and regulatory shifts Many hotels are now incorporating environmental sustainability practices (ESP), such as adopting energy-saving systems, minimizing waste, and using eco-friendly resources to lower their environmental footprint (Acampora et al., 2022) These initiatives are not only in alignment with global sustainability objectives but also respond to evolving customer expectations Notably, 83% of global travelers now prioritize eco-conscious practices when choosing accommodations (Booking.com,

While sustainability practices are critical for addressing environmental issues, their integration into hotel operations raises concerns regarding their influence on perceived service quality and guest satisfaction Previous studies have recognized service excellence as a fundamental component influencing customer experience and fostering long-term guest relationships (Parasuraman et al., 1985) However, sustainability initiatives, such as reduced laundry frequency or the use of alternative materials, may lead to concerns about service degradation if not effectively managed or communicated to customers (Chen & Peng, 2012; Robinot & Giannelloni, 2010) This creates a critical dilemma for hoteliers: how to balance sustainability initiatives with maintaining high- quality service that meets customer expectations Moreover, despite the growing interest in both SQ and ESP, existing literature tends to treat these two dimensions separately, leaving a gap in understanding their combined shape customer satisfaction (CS) and foster long-term loyalty (CL)

This concern is particularly relevant in the Vietnamese context, where the tourism and hospitality industries have witnessed substantial expansion As one of Asia’s most popular destinations, Vietnam recorded more than 18 million international tourist arrivals each year prior to the COVID-19 outbreak, with visitation numbers now gradually rebounding (VNAT, 2021) Key destinations like Ha Long Bay, Da Nang, and Phu Quoc face mounting environmental pressures due to over-tourism and inadequate sustainable practices In response, the Vietnamese government introduced the Strategy for Tourism Development in Vietnam to 2030, through Decision No 147/QD-TTg, issued on January 22, 2020 This strategy prioritizes sustainable tourism development based on green growth, emphasizing environmental protection and alignment with global sustainability trends It encourages hotels and tourism businesses to adopt eco- friendly practices to ensure long-term industry growth (Government of Vietnam, 2020)

However, despite these governmental efforts, sustainability adoption still tends to be more advanced in developed countries, where stronger institutions and greater environmental awareness foster more effective implementation (Lita et al., 2014) In contrast, developing nations like Vietnam often struggle with low consumer awareness and weak policy enforcement Similar trends have been observed in Malaysia, where hotels tend to prioritize service enhancement over environmental policies (Abdul Samad et al., 2008) In Vietnam, this pattern is also evident, as many hotels continue to prioritize enhancing facilities and customer service over sustainability initiatives While some businesses have integrated environmental practices into their operations, widespread adoption remains limited This reluctance is largely attributed to various structural and financial barriers According to Siti Nabiha et al (2011), the implementation of sustainability practices in the hospitality industry is influenced by several key factors, including regulatory frameworks, consumer demand for eco-friendly services, competitive market pressures, organizational environmental awareness, and corporate attitudes toward change Expanding on these insights, Nguyen (2023) highlights that despite the growing global emphasis on sustainable tourism, many hotels in Vietnam remain hesitant to fully embrace environmental sustainability practices The primary obstacles include high implementation costs, lack of standardized guidelines, and a perceived trade-off between environmental responsibility and service quality These concerns create uncertainty for hotel managers, who must balance meeting sustainability expectations with maintaining perceived service excellence A critical issue arises regarding how guests perceive such trade-offs: whether they associate ESPs with diminished service quality, or if they are willing to accept adjustments in services for the sake of environmental values This leads to an essential research need to examine customer perceptions of ESP and the interrelationships between ESP, SQ, and CS; especially whether ESPs influence CS directly, or whether they modify the effect of SQ on CS

Furthermore, customer responses to green practices may not solely depend on the practices themselves, but also on individual differences in environmental concern (EC) While some guests appreciate eco-initiatives as value-aligned efforts, others may remain indifferent or even skeptical, depending on their level of ecological awareness and personal values (Dunlap & Jones, 2002; Schultz, 2001) Despite its conceptual relevance, EC has rarely been incorporated into integrated models of service evaluation and sustainability in the hospitality literature, especially in emerging markets like Vietnam This presents an opportunity to investigate how EC may shape guest satisfaction and loyalty in the context of environmentally proactive hotels

Given these complexities, this study seeks to empirically assess the relationships among ESP, SQ, CS, and customer loyalty (CL) in Vietnam’s hospitality industry Using quantitative analysis and structural equation modeling (SEM), this research aims to provide robust empirical evidence on how ESP and SQ influence customer perceptions and long-term behavioral intentions The findings will contribute to existing literature by addressing a critical research gap and offering practical insights for Vietnamese hoteliers seeking to implement sustainability measures without compromising service excellence and customer satisfaction.

Research objective and research questions

Building upon the rationale above, this study aims to investigate how service quality (SQ) and environmental sustainability practices (ESP) affect customer satisfaction (CS) and customer loyalty (CL) in Vietnam’s hotel industry While previous research has tended to examine service and sustainability factors in isolation, this study integrates both dimensions to better understand how they collectively shape customer evaluations and behavioral responses, especially in markets where green certification is limited and sustainability practices are implemented voluntarily

In addition, the study incorporates Environmental Concern (EC) as a psychological factor that may influence how guests perceive and react to sustainability efforts Although EC has been discussed in environmental psychology and consumer behavior literature, its integration into hospitality-focused service models remains limited This study seeks to assess EC’s predictive role in shaping satisfaction and loyalty outcomes in the presence of green practices

Accordingly, this study is structured around the following objectives:

1 To investigate how service quality (SQ) and environmental sustainability practices (ESP) impact customer satisfaction and customer loyalty within hotel settings that implement green initiatives in Vietnam

2 To assess the influence of customers’ environmental concern on their satisfaction and loyalty, particularly in hospitality environments that emphasize sustainability

Based on these aims, the research put forward the following questions:

1 What are the effects of service quality and environmental sustainability practices on customer satisfaction and loyalty in hotels implementing green initiatives in Vietnam?

2 What is the role of environmental concern in influencing customer satisfaction and loyalty in hotels implementing environmental sustainability practices?

To answer these questions, the study employs structural equation modeling (SEM) to generate empirical insights on the interconnectedness of service delivery, sustainability practices, and attitudinal variables in the hospitality sector.

Research method

This study applies a quantitative methodology to empirically examine the interrelation among service quality (SQ), environmental sustainability practices (ESP), customer satisfaction (CS), and customer loyalty (CL) in Vietnam’s hotel sector

Data will be collected via a structured questionnaire distributed directly to hotel guests with diverse experiences in service quality and sustainability efforts The survey items are adapted from established measurement instruments in academic literature, ensuring reliable and accurate results To encompass a wide range of perspectives, convenience sampling will be employed to include various hotel categories and customer demographics, enhancing the comprehensiveness of the findings.

For the analytical stage, the research utilizes Partial Least Squares Structural Equation Modeling (PLS-SEM) via SmartPLS software to examine both direct and indirect relationships across variables In addition, descriptive statistics will be employed using SPSS to uncover trends in customer perceptions and behaviors This analytical design allows for a data-driven understanding of how service delivery and sustainability initiatives together influence customer satisfaction and loyalty within the context of Vietnam’s hospitality industry.

Contributions

This study contributes to both theoretical knowledge and practical applications in the field of hospitality sustainability by addressing key research gaps and providing insights into balancing service quality (SQ) and environmental sustainability practices (ESP)

This study contributes to the literature by extending prior research in three specific ways First, it integrates service quality and environmental sustainability practices into a unified framework, an area that has been largely overlooked in prior research While previous studies often examine SQ and ESP separately, this research investigates their combined effects on customer satisfaction (CS) and customer loyalty (CL), providing a more comprehensive understanding of how sustainability initiatives influence guest experiences in the hospitality industry

Second, this study expands knowledge by incorporating the consumer perspective on ESP, which remains relatively underexplored in the context of developing countries like Vietnam Given the increasing demand for eco-conscious travel options, this research sheds light on how sustainability measures shape customer perceptions, satisfaction, and loyalty, helping to bridge the gap between industry practices and consumer expectations

Finally, this study contributes to sustainability research by examining the moderating role of environmental concern in the relationship between SQ, ESP, CS, and CL This adds depth to the understanding of how customers with varying levels of environmental awareness react to sustainability initiatives in hotels

This study suggests potential practical implications for different stakeholders within the hospitality industry

Firstly, the integrated model developed in this research provides hotel managers with a clearer understanding of how environmental sustainability practices (ESP) and service quality (SQ) jointly influence customer satisfaction (CS) and loyalty (CL) By examining the interaction between these dimensions, the study helps managers make more informed decisions when designing and implementing sustainability initiatives This is especially important in contexts like Vietnam, where many hotel operators hesitate to adopt eco-friendly practices due to concerns that such changes may be perceived as a decline in service quality The findings offer guidance on how to balance sustainability efforts with maintaining guest expectations, suggesting that when well communicated and appropriately managed, ESP can coexist with high service standards

Secondly, the findings offer guidance on how to respond to changing customer expectations, especially among environmentally conscious guests As travelers with high environmental concern (EC) increasingly prefer sustainable hotels, managers can use these insights to craft communication and service strategies that resonate with different customer segments This helps build stronger connections with green-oriented guests, maintain satisfaction levels, and foster brand loyalty

This study offers valuable insights for policymakers and tourism authorities in Vietnam aiming to advance sustainable tourism The findings can guide the creation of regulatory frameworks and incentive programs that motivate hotels to implement environmentally sustainable practices, fostering long-term competitiveness and supporting the nation’s green growth objectives.

Scope of the Study

This study focuses specifically on three- to five-star hotels in Vietnam that have adopted environmental sustainability practices (ESP), regardless of their formal green certification status The decision to concentrate on this segment stems from the greater likelihood that such hotels possess the infrastructure, management capacity, and customer expectations necessary to implement and visibly communicate sustainability initiatives In contrast, lower-tier hotels (e.g., one- or two-star) may have more limited resources or informal practices, which are often not systematically communicated or standardized As a result, they were excluded to ensure alignment with the study’s objective of evaluating guest perceptions of formalized or visible sustainability efforts within structured service environments

The research targets the perceptions of hotel guests to examine how ESP and service quality (SQ) jointly influence their satisfaction and loyalty Data were collected through self-reported surveys administered to guests staying at environmentally proactive hotels The analysis centers on customers’ subjective experiences and evaluations The study is context-specific to Vietnam, where sustainability adoption in the hospitality sector is still evolving As such, the findings reflect the influence of cultural, regulatory, and environmental awareness conditions unique to this emerging tourism market.

LITERATURE REVIEW, THEORETICAL BACKROUND AND

Theoretical framework

The research is grounded in the Stimulus–Organism–Response (S-O-R) paradigm, which serves as the foundational theoretical lens for exploring the dynamics between service quality (SQ), environmental sustainability practices (ESP), and customer outcomes such as satisfaction (CS) and loyalty (CL) within Vietnam’s hotel sector Originally developed by Mehrabian & Russell (1974), the model suggests that external stimuli, like environmental or service-related factors, can influence individuals’ internal psychological or emotional states, which may subsequently shape their behavioral intentions and actions

The model comprises three key components: the stimulus (S), which includes external factors like ambient conditions, service encounters, or visible sustainability practices that trigger cognitive or emotional responses; the organism (O), representing internal psychological processes such as satisfaction, trust, or emotional arousal evoked by the stimulus; and the response (R), which indicates the behavioral outcome, including approach or avoidance behaviors, repurchase intentions, or customer loyalty (Jacoby, 2002; Jani & Han).

The S-O-R framework has been widely applied across marketing, retail, and service domains to understand how individuals process their environment before engaging in behavioral action In hospitality research, it has been particularly useful in explaining how service-related cues influence customers’ internal states and behavioral intentions For example, Su & Swanson (2017) applied the model to examine how destination social responsibility (stimulus) affected tourists’ emotional connection and environmentally responsible behavior (response), mediated by psychological attachment (organism) Similarly, Fan et al (2023) and Chang et al (2024) employed the framework to explore how green practices and service quality influence satisfaction and word-of-mouth in eco-friendly hotel contexts

Over time, scholars have extended the traditional S-O-R model in several directions to better capture the complexity of consumer behavior One common extension involves the inclusion of moderating variables, such as personality traits, environmental concern, or demographic characteristics, which affect directly the strength or direction of the relationships among S, O, and R (Su & Swanson, 2017) Another extension incorporates multiple organismic states, such as combining emotional and cognitive processes (e.g., satisfaction and brand trust), allowing for more layered interpretations of customer experience (Jani & Han, 2015)

Recent studies have also applied the S-O-R model in sustainability and green consumption contexts, particularly in eco-friendly hospitality services Here, sustainability cues such as green certifications, energy-saving devices, and environmental messaging function as stimuli that activate psychological evaluations (e.g., satisfaction, perceived responsibility), which then influence pro-environmental behaviors (Chang et al., 2024; Kapoor et al., 2021)

The S-O-R framework in this study is utilized by conceptualizing both Service Quality (SQ) and Environmental Sustainability Practices (ESP) as external stimuli (S) that shape hotel guests’ internal evaluation (O) in the form of Customer Satisfaction (CS) This internal state is expected to lead to Customer Loyalty (CL) as the behavioral response (R) Unlike prior studies that primarily examined indirect effects via satisfaction or other psychological mechanisms, this study also explores the possibility of direct pathways from SQ and ESP to CL, thereby providing a more well-rounded understanding of the way customers respond to service and sustainability cues Furthermore, the model is extended by incorporating Environmental Concern (EC) as an organism-based moderating variable, based on the premise that guests with stronger environmental values may process sustainability-related stimuli differently and exhibit varying levels of satisfaction and loyalty This approach enables the empirical testing of both mediated and moderated relationships, offering an enriched application of the S-O-R model in the sustainability-focused hospitality services context.

Literature review

Service quality (SQ) is widely recognized as a cornerstone concept in hospitality and service management research, as it has important role to play in hotel and service management research, especially in customer perceptions and satisfaction Within hospitality settings, service quality typically refers to the extent to which service performance aligns with or surpasses client expectation (Parasuraman et al., 1985)

Among various conceptualizations of service quality, the SERVQUAL model, which developed by Parasuraman et al (1985) and refined in 1988 remains the most widely adopted framework for evaluation This model delineates five core dimensions of service quality "Tangibles" describe visible features such as the hotel’s infrastructure, tools, and staff appearance “Reliability” reflects the consistency and precision with which a hotel delivers its promised services “Responsiveness” indicates how eager and quick employees are in assisting guests “Assurance” refers to staff expertise, politeness, and their capacity to build trust with customers Lastly, “empathy” relates to offering tailored and considerate service to meet the specific needs of each guest Collectively, these dimensions provide a comprehensive framework for evaluating perceived service quality across different hospitality settings

The relationship between service quality and customer satisfaction (CS) has been firmly established in both theoretical and empirical literature According to the Expectancy– Disconfirmation Theory (Oliver, 1980), satisfaction arises when customers perceive that actual service performance exceeds their initial expectations Within this framework, service quality acts as the external stimulus that initiates internal cognitive and emotional evaluations, resulting in satisfaction This is consistent with the Stimulus– Organism–Response (S-O-R) paradigm (Mehrabian & Russell, 1974), which posits that external service-related stimuli affect internal affective states, such as satisfaction, which in turn shape behavioral intentions

In this view, high service quality enhances customer satisfaction by signaling the hotel's capability, attentiveness, and commitment to fulfilling guest needs When hotel staff respond promptly and courteously, guests may experience both functional satisfaction and emotional fulfillment Grace & O’Cass (2005) and Ladhari (2009) argue that satisfaction in hospitality settings stems from both utilitarian performance and symbolic experience, such as trust and emotional comfort These emotional states amplify the overall evaluation of service quality, thereby increasing satisfaction

Empirical findings in hospitality research reinforce this relationship Juwaheer and Lee Ross (2003) found that reliable and responsive service enhances guest satisfaction in Mauritian hotels Similarly, Nguyen et al (2015) and Tuan & Linh (2014) showed that Vietnamese hotel guests highly value attentiveness and staff professionalism, which directly linked to satisfaction Based on this evidence, the following hypothesis is proposed:

H1: Service Quality (SQ) positively influences Customer Satisfaction (CS)

Beyond satisfaction, service quality also plays a critical role in shaping customer loyalty (CL), defined as a guest’s intention to return, recommend, and maintain long-term engagement with a hotel (Oliver, 1999) High service quality reinforces customer trust, reduces perceived risk, and fosters favorable emotional attachment, all of which contribute to repeat patronage and advocacy (Kandampully et al., 2015) Even when satisfaction is not explicitly measured, consistently high service performance can directly influence customer loyalty through habit formation and brand confidence (Caruana, 2002)

Research across multiple hospitality studies, including those in Vietnam (Nguyen et al., 2015; Tuan & Linh, 2014), confirms that service quality (SQ) directly influences customer loyalty (CL) This consistent validation leads to the hypothesis that enhancing SQ can significantly boost customer loyalty in the hospitality industry.

H3: Service Quality (SQ) positively influences Customer Loyalty (CL)

2.2.2 Environmental Sustainability Practices in Hotels

The increasing global concern regarding climate change, resource depletion, and environmental degradation has positioned environmental sustainability practices (ESP) as a critical element of responsible hotel management ESP in hotels signifies a collection of approaches and initiatives designed to mitigate environmental consequences via operational, policy, and behavioral interventions These practices are not only aligned with economic and social responsibility goals in the long run, but also contribute to the industry’s response to Sustainable Development Goals (SDGs) stated by the United Nations, notably SDG 12 (Responsible Consumption and Production) and SDG 13 (Climate Action)

Conceptually, sustainability practices are rooted in the triple bottom line framework, which emphasizes the need to harmonize environmental, social, and economic outcomes to ensure long-term organizational viability (Elkington, 1998) In the hospitality context, this often translates into the implementation of green practices, which are environmentally friendly business behaviors intended to reduce waste, conserve energy and water, and enhance environmental well-being (Joshi & Rahman, 2015)

Scholars have proposed various frameworks for categorizing ESP in hotels A widely cited classification by Baratta et al (2018) identifies five major dimensions of ESP: (1) waste management, (2) water conservation, (3) energy efficiency, (4) sustainable purchasing, and (5) people engagement This categorization reflects the increasing breadth of sustainability efforts, encompassing both physical infrastructure and stakeholder involvement

However, other researchers have proposed more condensed models For example, Kasimu et al (2012) grouped ESP into four categories: energy management, water savings, waste management, and general green support practices Meanwhile, Kim et al

(2012) developed a five-program framework based on operational focus areas such as room-based solid waste and water management, cleaning processes, biodiversity, and customer-facing water-saving options These differences highlight how ESP classifications may vary depending on regional contexts, hotel type, and research objectives Despite this variation, most models consistently include three fundamental categories - energy, water, and waste management - as central to ESP

Recognizing the need for standardized regional guidance, ASEAN Secretariat (2012) established the ASEAN Green Hotel Standard to promote sustainable hospitality in Southeast Asia This standard outline 11 key criteria, including energy and water efficiency, waste reduction, use of environmentally friendly products, community engagement, and environmental education for staff and guests This standard provides a rigorous benchmark for certification while guiding continuous improvement in sustainable hotel operations across ASEAN member states, including Vietnam

Numerous empirical studies have explored the relationship between ESP and key customer outcomes, particularly customer satisfaction (CS) and customer loyalty (CL), though these findings are not entirely consistent A growing body of research suggests that well-communicated and visibly implemented ESP can enhance guest satisfaction and foster positive behavioral intentions For example, Moise et al (2018) found that the adoption of green initiatives in Spanish hotels significantly improved guest satisfaction, which subsequently influenced revisit intentions and positive word-of- mouth Similarly, Merli et al (2019) demonstrated that guests at eco-labeled hotels in Italy reported higher satisfaction and loyalty, especially when sustainability practices aligned with their personal values

Some studies reveal that certain green practices in hotels can be misunderstood by guests, leading to negative outcomes For instance, Robinot & Giannelloni (2010) found that actions like removing complimentary single-use amenities were often seen as cost-cutting rather than sustainability efforts This misperception resulted in decreased guest satisfaction and heightened skepticism, particularly when sustainability messaging lacked clarity or was poorly communicated.

In addition, other studies highlight the conditional nature of these effects Moise et al

Research Gaps and Research Questions

Despite the growing body of literature on sustainable service delivery and customer behavior in hospitality, several important research gaps remain, particularly in emerging markets such as Vietnam, where sustainability transitions and customer expectations are still evolving

First, while both Service Quality (SQ) and Environmental Sustainability Practices (ESP) have been extensively examined in relation to Customer Satisfaction (CS) and Customer Loyalty (CL), the majority of previous studies have explored these variables separately There is a lack of integrated empirical models that examine how SQ and ESP jointly influence CS and CL within the same hospitality setting Especially in Vietnam, where green practices are increasingly adopted but often lack formal certification or widespread visibility, little is known about how these two constructs interact to shape customer experience

Second, existing research has established the positive effects of ESP on customer outcomes in developed economies (Merli et al., 2019; Moise et al., 2018) However, empirical findings remain mixed and context-dependent Some studies (Robinot & Giannelloni, 2010) suggest that green practices, if poorly communicated, may be misinterpreted as cost-cutting measures, leading to reduced satisfaction These findings highlight the need for context-specific investigations, especially in regions like Southeast Asia, where environmental awareness is still maturing and green signals may be undervalued or misunderstood Moreover, most prior studies have been conducted in developed regions with high environmental literacy In contrast, Vietnam represents a theoretical and empirical gap, where green initiatives are on the rise, but customer expectations and perceptions remain in flux Little is known about how Vietnamese hotel guests evaluate service quality and sustainability practices together, and how this interaction affects loyalty and satisfaction

Third, although the role of Environmental Concern (EC) as a moderating variable has been theoretically acknowledged in green consumer behavior and hospitality research (De Canio et al., 2021; Ham & Han, 2013), its application remains limited in integrated service models, particularly in emerging markets Most existing studies examine EC's moderating effect on a single relationship, such as between green practices and satisfaction (Han & Yoon, 2015), without testing EC as a cross-cutting moderator that shapes multiple paths in the customer evaluation process Moreover, EC is often treated as a peripheral or background factor, rather than as a value-based filter that conditions how customers cognitively and emotionally respond to both service quality and sustainability cues (Groening et al., 2018; White et al., 2019) In Vietnam, where environmental awareness is unevenly distributed and green hospitality is still developing, this gap becomes especially salient Limited studies have shown how EC moderates the relationship among core service attributes (e.g., SQ, ESP, CS) and behavioral outcomes (e.g., CL) in a unified model Addressing this gap will enhance theoretical understanding of value-driven segmentation and offer more tailored strategies for sustainable service design in culturally diverse and transitioning markets

Given these limitations, this study seeks to construct and empirically evaluate an integrated model that explores the interrelationships among Service Quality (SQ), Environmental Sustainability Practices (ESP), Environmental Concern (EC), Customer Satisfaction (CS), and Customer Loyalty (CL) within Vietnamese hotel sector By utilizing S-O-R framework, the study simultaneously examines direct, indirect (mediating), and moderating effects to offer a more comprehensive explanation of customer behavior in sustainability-oriented hospitality environments In doing so, it contributes to closing key theoretical and contextual gaps by broadening the use of the S-O-R paradigm within green service research in emerging countries

This project aims to solve the stated research gaps by answering the subsequent questions:

1 To what extent do service quality and environmental sustainability practices influence customer satisfaction and loyalty in Vietnamese hotel?

2 What is the role of Environmental Concern (EC) in shaping customer responses to green hotel practices in Vietnam?

Conceptual framework

Grounded in the Stimulus–Organism–Response (S-O-R) model (Mehrabian & Russell,

1974), this study conceptualizes Service Quality (SQ) and Environmental Sustainability Practices (ESP) as external stimuli that influence Customer Satisfaction (CS) (organism), which subsequently drives Customer Loyalty (CL) (response) The framework captures both direct and indirect pathways from SQ and ESP to CL, with CS acting as a mediator

In addition, Environmental Concern (EC) is incorporated as a moderating variable Specifically, EC is hypothesized to moderate the relationships between service attributes (SQ, ESP) and customer outcomes (CS, CL) It also moderates the link between CS and

CL By doing so, EC reflects individual differences in value orientation, shaping how customers process and respond to sustainability signals within the service environment (De Canio et al., 2021; Groening et al., 2018; Han & Yoon, 2015)

The proposed model is presented in Figure 2.1 and includes the following hypotheses:

H1: Service Quality (SQ) positively influences Customer Satisfaction (CS)

H2: Environmental Sustainability Practices (ESP) positively influence Customer

H3: Service Quality (SQ) positively influences Customer Loyalty (CL)

H4: Environmental Sustainability Practices (ESP) positively influence Customer

H5: Environmental Sustainability Practices (ESP) moderate the relationship between

Service Quality (SQ) and Customer Satisfaction (CS)

H6a: Environmental Concern (EC) moderates the relationship between Environmental

Sustainability Practices (ESP) and Customer Satisfaction (CS)

H6b: Environmental Concern (EC) moderates the relationship between Environmental

Sustainability Practices (ESP) and Customer Loyalty (CL)

H6c: Environmental Concern (EC) moderates the relationship between Service Quality

(SQ) and Customer Satisfaction (CS)

H6d: Environmental Concern (EC) moderates the relationship between Service Quality

(SQ) and Customer Loyalty (CL)

H6e: Environmental Concern (EC) moderates the relationship between Customer

Satisfaction (CS) and Customer Loyalty (CL

H7a: Customer Satisfaction (CS) mediates the relationship between Service Quality

(SQ) and Customer Loyalty (CL)

H7b: Customer Satisfaction (CS) mediates the relationship between Environmental

Sustainability Practices (ESP) and Customer Loyalty (CL).

RESEARCH METHODOLOGY

Research Approach

The thesis adopts a method of quantitative analysis to test proposed hypotheses within a structured conceptual framework The method is suitable for identifying causal relationships among latent constructs and examining the impacts of service quality, environmental sustainability practices, and environmental concern on customer satisfaction and loyalty Due to the intricate nature of the study model, Partial Least Squares Structural Equation Modeling (PLS-SEM) was used to estimate the measurement and structural components of the model

Population and Sampling Strategy

The target population comprises hotel guests in Vietnam who have experienced environmental sustainability initiatives during their stay The sampling frame focused on guests staying in mid-range to five-star hotels, where sustainability practices are more likely to be adopted

To improve data accuracy and ensure relevance to the research objectives, a set of screening questions was placed at the beginning of the questionnaire Respondents were required to confirm that they had stayed at a hotel in Vietnam within the past twelve months and could clearly recall and evaluate their experience at a three-star hotel or above These establishments are more likely to have visible environmental initiatives in place, such as towel reuse programs, waste separation bins, or the use of eco-friendly products Respondents were also asked to reflect on a specific hotel stay to ensure that their responses referred to a distinct and recent lodging experience

The research utilized a non-probability convenience sampling approach, which is appropriate for exploratory studies involving complex models and latent constructs, especially when a formal sampling frame is unavailable Participants were recruited online through social media travel groups, university networks, and direct referrals Platforms such as Facebook and Zalo were used to disseminate the survey link, reaching a diverse group of respondents across major regions of Vietnam, including Northern, Central, and Southern provinces

To qualify for inclusion, participants needed to be Vietnamese residents aged 18 or older, with a hotel stay in Vietnam within the previous year, and a level of familiarity sufficient to assess both the service quality and the environmental practices of the hotel Only guests who had stayed at three-star hotels or higher were included in the final sample, as these are more likely to adopt formal or informal sustainability practices On the other hand, individuals were excluded if they had not stayed in a hotel within the past year, had only stayed in budget accommodations with no identifiable green initiatives, or were unable to recall key aspects of their lodging experience relevant to the research variables.

Sampling size

In structural equation modeling (SEM), especially when using the partial least squares (PLS) approach, determining an adequate sample size is essential to ensure the reliability, statistical power, and generalizability of the findings While there is no absolute rule for determining sample size, various guidelines and empirical recommendations exist based on model complexity and the nature of relationships being tested

According to Hair et al (2019), models that incorporate multiple constructs, mediation and moderation effects, or higher-order constructs (HOC) typically require a minimum sample size of 200 to 300 observations This recommendation ensures sufficient statistical power to detect meaningful effects and to produce stable and reliable parameter estimates The presence of complex relationships, such as indirect and interaction effects, further increases the required sample size due to higher estimation demands

In this study, the research model includes a relatively large number of constructs and integrates both mediation and moderation effects, along with higher-order constructs Therefore, following the guidelines suggested by Hair et al (2019), a minimum threshold of 300 participants was set to ensure methodological rigor Ultimately, the study collected responses from 354 valid participants, which exceeds the recommended threshold and supports the robustness of the analysis.

Data collection process

Designed to gather guest opinions of service quality and environmental sustainability policies in Vietnamese hotels, the structured, self-administered online questionnaire used for the thesis English was used for development of the questionnaire; it was subsequently translated into Vietnamese and polished using a back-translation technique for ensuring semantic coherence and clarification

Ten people were used for a pilot test to examine the phrasing, logic, and clarity of the questionnaire items before starting the full-scale survey Although the pilot did not involve formal data collection, respondents were invited to provide feedback on potential ambiguities, cultural nuances, and comprehension issues Based on their input, minor adjustments were made to improve the readability of the survey These pilot responses were not included in the final dataset

The finalized survey instrument included an introductory statement, followed by screening questions designed to verify respondent eligibility based on criteria such as age, recent hotel experience in Vietnam, and the ability to evaluate hotel services and environmental initiatives Only qualified participants were allowed to proceed further

The survey was distributed over a period of approximately one-month period, from February 25 to March 23, using non-probability convenience sampling, primarily through social media platforms (e.g., Facebook, Zalo travel groups), university mailing lists, and personal networks These channels provided access to a diverse pool of Vietnamese hotel guests who had stayed in three-star or higher-rated properties

Respondents were informed their data would be used only for this study; participation was fully anonymous No personally identifiable information was collected, ensuring full compliance with ethical standards for human subject’s research

By the end of the survey, 392 responses in all had already been collected A data screening process was then applied to ensure data quality and alignment with the study's objectives At first, 17 responses were eliminated due to the fact that those who responded had not stayed in a hotel within the past two years, which is a fundamental requirement for participation Among the remaining 375 responses, 21 were excluded due to respondents indicating stays in homestays or 1- to 2-star hotels, or for providing invalid answers regarding hotel classification This study excludes these types of accommodations, as it concentrates on customer perceptions of service quality and environmental sustainability practices in 3- to 5-star hotels After the cleaning process,

354 valid responses were retained for further analysis This final sample size meets the inclusion criteria and exceeds the minimum threshold recommended for complex models in PLS-SEM, thus providing a robust basis for the proposed research model.

Measurements

To ensure content validity and contextual relevance, all components were assessed on multi-item measures developed based on previously validated instruments The measurement items were revised and localized through translation and expert feedback to suit the Vietnamese hotel setting Responses have been recorded on a five-point Likert scale, where 1 refers to Strongly Disagree and 5 signifies Strongly Agree

Service quality was operationalized as a second-order reflective construct, measured across five first-order dimensions based on the SERVQUAL model proposed by Parasuraman et al (1988): “Tangibles” (4 items), “Reliability” (4 items),

“Responsiveness” (3 items), “Assurance” (4 items), and “Empathy” (3 items), resulting a total of 18 items The scale was adapted from Parasuraman et al (1988) and contextualized using items from Nguyen et al (2015) to reflect the Vietnamese hotel industry

ESP was measured using nine items amended and modified from Merli et al (2019) The measures capture guest perceptions of hotel sustainability initiatives in five categories: energy efficiency, water conservation, waste reduction, green purchasing, and environmental communication These dimensions were aligned with the ASEAN Green Hotel Standard to ensure policy relevance and regional applicability

Customer satisfaction had been measured with 4 items, based on model of Moise et al

(2018) The components reflect guests’ affective evaluation of their overall experience at the hotel, including fulfillment of expectations and emotional satisfaction

Customer loyalty had been evaluated with three items derived from Priyo et al (2019) The items reflect behavioral intents including likelihood of revisiting, brand preference, and recommendations through word-of-mouth

Environmental concern was accessed using 4 items derived from De Canio et al (2021) These items reflect the respondent’s general awareness of environmental issues, perceived responsibility, and value orientation toward ecological sustainability

Table 3.1 presents the complete list of measurement items, including their original sources and item codes Prior to conducting the primary analysis, the reliability and validity of all scales were assessed through a comprehensive series of measurement model evaluations within PLS-SEM.

Table 3.1 Measurement scales for all variables

1 = Strongly disagree to 5 = Strongly agree

“The hotel’s facilities and equipment are modern and well-maintained”

Nguyen et al (2015); Parasuraman et al (1988)

“The hotel environment is clean, and provides a comfortable atmosphere”

“The staff have a neat appearance that reflects professionalism”

“The hotel provides visually presented brochures and directories”

5 RE1 “The hotel delivers services on time as promised” 1 2 3 4 5

6 RE2 “The hotel staff handles guest requests with accuracy” 1 2 3 4 5

7 RE3 “The hotel sincerely solves problems when they arise” 1 2 3 4 5

“The hotel provides accurate information through its website or over the phone”

“Hotel staff are always willing to help whenever assistance is needed”

10 RES2 “The hotel staff respond to guests’ requests promptly” 1 2 3 4 5

“The hotel staff are flexible in meeting guests’ specific needs and demands”

12 AS1 “I feel safe and secure during my stay at the hotel” 1 2 3 4 5

“The hotel staff demonstrate deep knowledge to answer my questions about their services and products”

“The hotel staff are always polite and courteous when providing services”

“The hotel staff possess professional occupational skills occupational skills”

16 EP1 “Hotel staff provides guests individual attention” 1 2 3 4 5

17 EP2 “Hotel staff understand guests’ specific needs” 1 2 3 4 5

“Hotel staff show a positive attitude when receiving guest feedback”

II Environmental sustainability practices (ESP) (9 items)

“The hotel implements energy saving practices (e.g., automatic lights switching off, energy-saving lighting device, key cards,…)”

Adopt and modified from Merli et al (2019)

“The hotel implements water saving practices (e.g., the hotel encourage guest to ask for new linen only when necessary)”

“The hotel tries to avoid disposable or single-dose products (e.g., using refillable dispensers for shampoo, soap, glass bottles, fabric bags, )”

The hotel provides clearly labeled and convenient waste sorting bins (e.g., for recyclables, organic waste, general waste, etc.)

“The hotel prioritizes the use environmental certified or green label products (e.g., toiletry products, paper,)”

Organic or locally sourced food and beverages are widely used in restaurants or the hotel minibar

“The hotel informs the guests about the good environmental practices implemented”

“The hotel provides its guests with information on how they can contribute to reduce the hotel’s environmental impact”

“The hotel provides its guests with information on the environmental and cultural activities available in the area”

III Customer satisfaction (CS) (4 items)

28 CS1 “The hotel stay was exactly what I needed” 1 2 3 4 5

Adopt and modify from Moise et al

29 CS2 “I am satisfied with my decision to stay at this hotel” 1 2 3 4 5

30 CS3 “Choosing this hotel was a wise decision” 1 2 3 4 5

31 CS4 “Overall, staying at this hotel was a good experience” 1 2 3 4 5

IV Customer loyalty (CL) (3 items)

32 CL1 “I tend to prefer this hotel to the other hotels” 1 2 3 4 5

33 CL2 “I intend to re-visit this hotel in the future” 1 2 3 4 5

34 CL3 “I am willing to recommend this hotel to other” 1 2 3 4 5

“If all of us, individually, contributed to environmental protection, it would have a significant effect”

“Everyone is responsible for protecting the environment in their everyday life”

“Preserving and protecting the environment should be one of our priorities”

“We should take responsibility for environmental issues, as we are the cause of environmental damage”

Specification of Higher- Order Construct (HOC)

This study adopts a reflective-reflective hierarchical component model (HCM) to conceptualize Service Quality (SQ), grounded in the widely accepted SERVQUAL framework introduced by (Parasuraman et al., 1988) This higher-order model includes two hierarchical levels: five first-order constructs (lower-order constructs, LOCs), which are Tangibles (TA), Reliability (RE), Responsiveness (RES), Assurance (AS), and Empathy (EP), and one second-order construct (higher-order construct, HOC), namely Service Quality (SQ)

The reflective-reflective model is theoretically justified based on the understanding that these five LOCs reflect a broader latent concept of service quality These dimensions are expected to be highly interrelated, as they represent various but interdependent aspects of the same underlying construct (Hair et al., 2021)

Given the complexity of the hierarchical structure and the nature of the constructs, this research employs PLS-SEM via SmartPLS 4 software (Becker et al., 2015) The use of this PLS-SEM is particularly suitable for estimating complex models involving latent variables and hierarchical constructs, and it does not require multivariate normality assumptions, making it appropriate for exploratory and predictive research (Hair et al.,

A two-stage analytical technique was employed to evaluate the measurement model (Becker et al., 2012; Hair et al., 2021): i Stage One – Evaluation of the First-order Measurement Model: Initially, each first-order construct (LOC) is assessed independently using standard reliability and validity criteria This includes assessing indicator loadings, composite reliability (CR), Cronbach’s Alpha, Average Variance Extracted (AVE) ii Stage Two – Higher-order Measurement Model Assessment: After establishing reliability and validity in the first stage, Stage Two involves creating latent variable scores (LVS) for each validated LOC These LVS are computed within SmartPLS and then utilized as reflective indicators of the second-order construct (SQ) In this approach, the higher-order construct is modeled as reflectively indicated by these LVS, reflecting the theoretical understanding that SQ is a broader latent concept collectively represented by these five dimensions The subsequent analysis in Stage Two involves reassessing internal consistency reliability, convergent validity, and discriminant validity of the higher-order construct

This approach enables a comprehensive assessment of both the individual dimensions of SQ and the overarching construct, while minimizing model complexity and multicollinearity The use of SmartPLS further facilitates this process through its robust handling of hierarchical models and latent variable estimation.

DATA ANALYSIS RESULTS

Demographic characteristics

4.1.1 Demographic Characteristics of the Respondents

Table 4.1 summarizes the demographic profiles of the 354 valid participants who met the screening criteria

Table 4.1 Profile background of the respondents

The sample's gender distribution, which includes 354 valid replies, is comparatively balanced, with 54.2% of respondents being female, 44.9% being male, and 0.8% choosing not to reveal their gender This indicates a slightly higher participation rate from female customers in the hospitality industry

The majority of respondents were aged 25 to 35 years (44.1%), followed by those under

25 years (27.7%) and those aged 36 to 45 years (21.7%) Older respondents aged 46–60 years and over 60 years accounted for only 5.4% and 1.1%, respectively This suggests that the sample was largely composed of young to middle-aged adults, reflecting a demographic that is more likely to travel and engage with hotel services

Regarding income, approximately 58.5% of respondents earned below 20 million VND per month, with 35.3% earning between 10 and under 20 million VND, and 23.2% earning below 10 million VND Meanwhile, 24.9% had incomes between 20 and under

30 million VND, and a smaller group (16.6%) earned over 30 million VND/month These figures reflect a sample with relatively modest to moderate income levels, which may influence their expectations and satisfaction in hotel service quality and environmental practices

As for educational attainment, 55.9% of respondents held a university degree and 25.7% had completed postgraduate education Only 18.4% had college-level education or below The relatively high level of education among respondents indicates that most participants were likely to be well-informed and potentially more sensitive to sustainable and service quality issues

4.1.2 Hotel Profile and Stay characteristics

About Hotel Profiles and Stay Characteristics, Table 4.2 summarizes these characteristics of the hotels and customer stay profiles

Table 4.2 Characteristics of Hotel’s Profile and customer stay profiles

Other purposes: visiting friends, religious activities, sporting events, eco-tourism, etc)

Frequency of stay in a year

In terms of hotel category, 44.6% of respondents stayed at five-star hotels, 31.7% in four-star hotels, and 23.7% at three-star lodgings This signifies that most of the sample encountered premium accommodation services, which aligns with the research focus on quality and sustainability

Geographically, 48.3% of respondents stayed at hotels in Northern Vietnam, 31.4% in the South, and 20.3% in the Central region This distribution provides a well-diversified geographical coverage, allowing the study to reflect variations in regional hotel services

As for the purpose of travel, the dominant reason was leisure or relaxation (63.3%), followed by business (22%), and a smaller percentage for educational purposes (3.7%) and others (11%) such as eco-tourism, visiting relatives, or spiritual travel These statistics show that most guests viewed their hotel stay as a leisure activity, which may impact how they perceive service quality and sustainability

Regarding duration, 61.0% of respondents stayed 1–2 nights, and 34.2% stayed 3–5 nights, indicating that short-term stays dominate Furthermore, 68.1% of respondents reported staying in hotels once per year, and only 8.5% stayed more than 3 times a year, highlighting that for most, hotel stays are occasional rather than frequent.

Measurement test

To assess the internal consistency of the measurement instruments, the study utilized both Cronbach’s Alpha coefficients and item-total correlation scores Following the guideline proposed by DeVellis (2017), an Alpha coefficient equal to or above 0.70 is generally considered acceptable, while item-total correlation values exceeding 0.30 are viewed as indicating adequate reliability at the item level

All latent constructs demonstrated strong internal consistency, with reliability scores exceeding the 0.70 threshold—ranging from 0.847 for Empathy to 0.940 for Environmental Sustainability Practices Corrected item-total correlations for each item also surpassed 0.30, confirming item-level reliability Consequently, no items were removed, and all indicators remained for subsequent analyses.

Table 4.3 Reliability and Descriptive Statistics of Constructs

Regarding descriptive statistics, the mean scores for the five dimensions of SQ were relatively high, ranging from 3.93 (Empathy) to 4.13 (Assurance) Among these dimensions, Assurance (Mean = 4.13) and Reliability (Mean = 4.09) recorded the highest evaluations, suggesting that guests highly valued the reliability and trustworthiness of hotel services

For the other constructs, the mean scores varied slightly Environmental Sustainability Practices (ESP) received a moderate evaluation (Mean = 3.56), indicating room for improvement in hotels' green initiatives In contrast, Customer Satisfaction (CS) and Environmental Concern (EC) were rated relatively high (Mean = 4.05 and 4.31, respectively), reflecting positive guest attitudes toward their hotel experience and environmental awareness Meanwhile, Customer Loyalty (CL) showed a slightly lower mean score (Mean = 3.89), suggesting that while guests were generally satisfied, their loyalty may require stronger reinforcement strategies

The standard deviations ranged from 0.76 to 0.99, with higher variability observed in ESP and CL, indicating greater dispersion in guests' perceptions toward sustainability practices and loyalty

Overall, these results suggest that guests were generally satisfied with service quality and environmentally sustainable efforts, although there is room for improvement, particularly in enhancing loyalty and consistently promoting green practices across different hotels

To identify the underlying structure of the constructs, the study employed Exploratory Factor Analysis (EFA) using the Principal Component Analysis (PCA) method with Varimax rotation Before conducting EFA, the suitability of the dataset was assessed by calculating the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and performing Bartlett’s Test of Sphericity As shown in Table 4.4, the KMO value reached 0.944, which is well above the commonly accepted threshold of 0.60 (Kaiser, 1974), indicating that the sample was appropriate for factor analysis Furthermore, Bartlett’s Test produced a highly significant result (χ² = 9858.003, df = 465, p < 001), providing statistical evidence that the variables were sufficiently intercorrelated to proceed with EFA

Table 4.4 KMO and Bartlett’s Test

Initially, all 31 observed items from independent variables were included in the analysis The results revealed a three-factor solution based on eigenvalues greater than 1, fulfilling Kaiser’s criterion (Kaiser, 1960) As shown in Table 4.5, these three components jointly explained 66.481% of the total variance after rotation, which exceeds the 50% threshold recommended for social science research (Hair et al., 2019)

This demonstrates a strong factor structure and ensures that the extracted components adequately represent the observed variables

An examination of the Rotated Component Matrix (Table 4.6) revealed that most items loaded distinctly onto their respective factors, displaying clear discriminant validity However, item TA1 presented cross-loading values, loading primarily on Component 1 (0.634) and to a lesser extent on Component 3 (0.522) Although the difference in cross- loading was relatively small (0.112), TA1 was still retained based on several strengths Firstly, TA1 exhibited a high corrected item-total correlation (0.775), reflecting strong internal consistency with the Tangibles dimension Secondly, reliability analysis indicated that removing TA1 would result in a slight decrease in Cronbach’s Alpha from 0.882 to 0.836, yet it would remain comfortably above the acceptable threshold of 0.70 (Nunnally, 1978) Finally, conceptually, TA1 aligns well with the theoretical construct it represents, offering important theoretical coherence and practical significance for subsequent analyses Therefore, considering these statistical and conceptual justifications, the decision was made to retain TA1 within the measurement model, and conclusively TA1 assigned to Component 1

These results collectively affirm the robustness and adequacy of the factor structure, justifying progression to Structural Equation Modeling (SEM) in subsequent stages of analysis.

Correlation analysis

To further assess the relationships among constructs prior to the structural model evaluation, Pearson correlation analysis was conducted Table 4.7 presents the correlation matrix among the five main constructs: SQ, ESP, EC, CS, CL

Table 4.7 Correlation matrix among constructs

The results revealed that all constructs were significantly and positively correlated at the 0.01 level (two-tailed) The correlation coefficients ranged from 0.416 to 0.852, indicating moderate to strong relationships among the constructs

Specifically, Service Quality (SQ) exhibited strong positive correlations with CS (r 0.803, p < 001) and CL (r = 0.698, p < 001), suggesting that enhanced perceived service quality correlates with increased satisfaction and loyalty ESP also demonstrated positive correlations with CS (r = 0.668, p < 001) and CL (r = 0.630, p < 001)

Importantly, no correlation coefficient exceeded 0.85, suggesting that multicollinearity is not a substantial issue among the constructs (Kline, 2015) Thus, the dataset is appropriate for subsequent structural model assessment.

Measurement model evaluation

Following the assessment of scale reliability, exploratory validity, and correlation analysis, this section proceeds to evaluate the measurement model through Confirmatory Factor Analysis (CFA) using PLS- SEM approach This evaluation aims to confirm the psychometric properties of all constructs before proceeding to test the structural relationships among them

SQ is conceptualized as a second-order reflective-reflective hierarchical construct composed of five key dimensions: TA, RE, RES, AS, and EP The measurement model is rigorously evaluated at both first-order and second-order levels to ensure robust validity and reliability This approach highlights the multi-dimensional nature of SQ, emphasizing its complex structure and the importance of each dimension in shaping overall quality Proper assessment at both levels allows for a comprehensive understanding of how these dimensions interact, providing valuable insights for researchers and practitioners in the field.

4.4.1 Evaluation of the First-Order Measurement Model

The outer loadings of all indicators were first examined Table 4.8 demonstrates that all items exhibited significant and robust loadings on their corresponding constructs, with standardized loadings surpassing the advised threshold of 0.70 (Hair et al., 2019) This indicates that individual indicators adequately represent their respective latent constructs The loadings for the five Service Quality dimensions ranged from 0.809 to 0.905, while those for ESP, EC, CS, and CL similarly demonstrated strong indicator performance No item needed to be removed

Table 4.8 Measurement model evaluation results

Constructs/ Indicators Mean St.dev Loading

The internal consistency of the measurement model was assessed using both Cronbach’s Alpha and Composite Reliability (CR) indicators As presented in Table 4.8, all constructs achieved Alpha and CR scores above the threshold of 0.70, confirming that the items demonstrated satisfactory levels of internal reliability (Hair et al., 2021) To evaluate convergent validity, the study applied the Average Variance Extracted (AVE) criterion Results showed that all constructs had AVE values exceeding 0.50, indicating that a substantial proportion of variance in observed items was explained by their corresponding latent variables (Fornell & Larcker, 1981) The AVE scores ranged between 0.677 for Environmental Sustainability Practices (ESP) and 0.806 for Customer Loyalty (CL), thereby supporting the presence of convergent validity across all measured constructs

Figure 4.1 First – Order Measurement Model

4.4.2 Evaluation of the Second-Order Measurement Model

The second-order measurement model focuses on Service Quality (SQ) as a reflective- reflective higher-order construct, composed of the five lower-order constructs (TA, RE,

RES, AS, EP), and is evaluated based on the PLS-SEM guidelines (Becker et al., 2012; Hair et al., 2021) As shown in Table 4.9, all lower-order constructs loaded strongly onto the second-order construct, with loadings above 0.89, confirming their substantial contribution to the overarching SQ construct

Table 4.9 Outer Loadings of High Order Constructs

Then, Composite Reliability (CR) and AVE score were assessed for SQ as a second- order construct As shown in Table 4.10, the second-order construct SQ achieved a Cronbach’s Alpha of 0.954 and a CR of 0.965, both of which exceed the recommended threshold of 0.70, indicating excellent internal consistency (Hair et al., 2021) Additionally, the average variance extracted (AVE) was 0.846, exceeding the minimum criterion of 0.50, confirming the presence of convergent validity (Fornell & Larcker, 1981; Hair et al., 2019)

Table 4.10 Construct reliability and validity for Second – order construct

Discriminant validity was assessed using the Heterotrait-Monotrait (HTMT) ratio All HTMT values between the second-order construct (SQ) and the first-order constructs

(ESP, EC, CS, CL) were below the conservative threshold of 0.90 (Henseler et al.,

2015), except for the HTMT between CS and CL, which slightly exceeded the threshold at 0.962 Although this value is marginally above the recommended threshold, it remains conceptually acceptable given the theoretical distinction between the two constructs Customer satisfaction (CS) reflects a customer’s evaluative and affective response to a specific service encounter, whereas customer loyalty (CL) captures a broader behavioral intention and long-term commitment toward the service provider (Dick & Basu, 1994; Oliver, 1980) Despite their conceptual separation, the two constructs are empirically interrelated, particularly in hospitality settings, where prior studies have consistently demonstrated a strong correlation between satisfaction and loyalty (Cronin & Taylor, 1992; Kandampully & Suhartanto, 2000).Therefore, this minor deviation does not compromise the overall discriminant validity of the model

Table 4.11 HTMT Matrix for Discriminant Validity Assessment

Finally, to further validate the hierarchical component model, bootstrapping with 5,000 resamples was conducted All path coefficients from the second-order construct (SQ) to its five lower-order dimensions (TA, RE, RES, AS, and EP) were statistically significant at p < 0.001 This provides strong empirical support for the reflective-reflective structure of the Service Quality construct

The measurement model demonstrates strong indicator reliability, internal consistency, convergent validity, and discriminant validity, confirming its reliability and robustness This solid foundation allows for accurate assessment of the structural model in subsequent analysis stages.

Assessment of the Structural Model

To assess the structural model, a few diagnostic criteria were employed, including collinearity assessment (VIF), coefficient of determination (R²), predictive relevance (Q²), effect size (f²), model fit indices, and path coefficients (significance and relevance) This follows the standard evaluation procedure for PLS-SEM proposed by (Hair et al.,

Before analyzing the structural relationships, collinearity among the predictor components was evaluated using VIF indicator According to (Hair et al., 2021), VIF values below 5 are generally considered acceptable, though values up to 10 may still be tolerated without serious multicollinearity concerns (Becker et al., 2015)

Most VIF values in the study stayed below the threshold of 5, indicating acceptable multicollinearity levels However, some indicators like ESP8 (VIF = 5.916), RE (VIF = 5.582), RES (VIF = 5.502), and ESP7 (VIF = 5.002) slightly exceeded this cutoff Despite this, none of the VIF values reached the critical level of 10, suggesting that collinearity is unlikely to significantly affect the accuracy of the path coefficient estimates.

Although model fit assessment is not mandatory in PLS-SEM, recent methodological advances recommend the inclusion of global fit indices to provide a comprehensive judgment of model adequacy (Hair et al., 2021) The results are presented in Table 4.13

Among the fit indices, the Standardized Root Mean Square Residual (SRMR) is widely regarded as the most appropriate global indicator in PLS-SEM, as it reflects the average discrepancy between the observed and predicted correlations In this study, the SRMR value of 0.061 falls below the recommended threshold of 0.08 (Henseler et al., 2015), indicating a good model fit

The Normed Fit Index (NFI), which compares the fit of the proposed model to that of a null (independence) model, yielded a value of 0.828 Since values above 0.80 are generally considered acceptable (Bentler & Bonett, 1980), this result suggests that the proposed model offers a substantial improvement over the baseline model in capturing the data’s variance–covariance structure

Discrepancy-based fit indices, d_ULS and d_G, measure the distance between the empirical and model-implied correlation matrices, with lower values indicating better fit In this study, d_ULS (1.19) and d_G (0.814) values support the model’s adequacy, aligning with the common preference for lower fit index scores based on recommendations by Hair et al (2021) Although PLS-SEM does not specify strict cutoff values for these indices, the obtained scores suggest a good model fit.

Taken together, these results confirm that the structural model demonstrates an acceptable level of model-data fit, supporting the overall adequacy of the theoretical model specified in this study

4.5.3 Endogenous Construct Assessment and Path coefficients

The structural model was evaluated following the standard procedure outlined by (Hair et al., 2013) This includes examining the explanatory power (R²), predictive relevance (Q²), effect sizes (f²), and hypothesis testing via the analysis of standardized path coefficients (β) and their significance levels obtained through a bootstrapping procedure

The coefficient of determination (R 2 ) was employed to evaluate the endogenous constructs' explanatory power R² signifies the proportion of variance in endogenous constructs that is explicable by their corresponding exogenous constructs R² values vary between 0 and 1, where elevated values signify greater explanatory power of the model According to (Hair et al., 2013), R² values of 0.75, 0.50, and 0.25 can be interpreted as substantial, moderate, and weak, respectively, in terms of predictive accuracy Table 4.14 indicates that the R² value of 0.760 for Customer Loyalty (CL) demonstrates that 76.0% of the variance in customer loyalty is accounted for by the antecedent constructs in the model, reflecting a significant level of explanatory power Customer Satisfaction (CS) exhibits a R² of 0.719, indicating that 71.9% of its variance is explained by its predictors The results demonstrate significant predictive capability for both constructs

The predictive relevance was evaluated using Stone-Geisser’s Q² statistic, derived from the blindfolding procedure Q² values exceeding zero signify that the model possesses predictive relevance for a particular endogenous construct (Geisser, 1974) The Q² values for CS (0.696) and CL (0.552) are both greater than zero, indicating that the model demonstrates strong predictive validity

Furthermore, effect size (f²) was calculated to assess the individual impact of each exogenous construct on the R² of the endogenous variables According to Cohen (2013) f² values of 0.02, 0.15, and 0.35 are considered small, medium, and large effects, respectively The effect of Service Quality (SQ) on CS was large (f² = 0.363), while the effect of ESP on CS was moderate (f² = 0.171) Other effects were either small or negligible

As presented in Table 4.14, the path from Service Quality (SQ) to Customer Satisfaction (CS) was positive and highly significant (β = 0.486, t = 7.360, p < 0.001), with a large effect size (f² = 0.363), confirming H1 This indicates that improvements in perceived service quality strongly contribute to enhanced customer satisfaction in the hotel context

Similarly, Environmental Sustainability Practices (ESP) also had a positive and significant effect on CS (β = 0.288, t = 6.338, p < 0.001), supporting H2 The associated effect size (f² = 0.171) suggests a moderate impact, emphasizing the relevance of sustainable practices in shaping customer perceptions of satisfaction

The direct impact of Service Quality (SQ) on Customer Loyalty (CL) was found to be statistically insignificant (β = 0.070, p = 0.265), indicating that service quality by itself does not directly drive customer loyalty This suggests that other mediating or moderating factors may play a crucial role in translating service quality into customer loyalty, highlighting the complexity of the relationship between these variables.

CL path (H4) showed a positive coefficient (β = 0.093) and was marginally significant at the 10% level (p = 0.052), indicating limited but noteworthy support for the direct influence of environmental sustainability on loyalty

Table 4.14 Structural path coefficients and endogenous construct assessment

Among the moderating hypotheses, the interaction effect of ESP × SQ on CS (H5) was marginally significant (β = -0.087, t = 1.743, p = 0.081), suggesting that the positive impact of service quality on satisfaction may be slightly diminished under conditions of high sustainability practice implementation Similarly, the interaction effect of EC × SQ on CS (H6c) was also marginally supported (β = -0.063, p = 0.097), hinting at a possible weakening of the SQ-CS relationship when customers are highly environmentally concerned

Evaluation of Mediation Effects

The final step in assessing the structural model involves testing the mediation effects, with a specific focus on the role of Customer Satisfaction (CS) as a mediator between the independent variables - Service Quality (SQ) and Environmental Sustainability Practices (ESP), and the dependent variable - Customer Loyalty (CL) Mediation analysis helps clarify whether the relationship between an exogenous and an endogenous construct is significantly transmitted through a third (mediating) variable In line with current best practices in PLS-SEM, this study adopts the bootstrapping approach proposed by Preacher & Hayes (2008) and Zhao et al (2010), which has been widely recognized for its robustness compared to earlier methods such as the Sobel test or the Baron & Kenny (1986) causal steps approach, both of which have been criticized for low statistical power and overly strict assumptions

Following the nonparametric resampling method, 5,000 bootstrap iterations were conducted to estimate the significance of the indirect effects The indirect path from SQ to CL through CS was found to be statistically significant (β = 0.388, p < 0.001), while the direct effect of SQ on CL was not significant (β = 0.070, p = 0.265), suggesting a situation of full mediation This means that the influence of service quality on customer loyalty occurs entirely through the perceived satisfaction of the customer In contrast, the indirect effect from ESP to CL via CS was also significant (β = 0.230, p < 0.001), but in this case, the direct effect remained marginally significant (β = 0.093, p = 0.052), implying the presence of partial mediation Hence, while ESP contributes to loyalty both directly and indirectly through satisfaction, SQ’s impact is entirely mediated by CS

To further assess the strength of the mediation effect, the Variance Accounted For (VAF) was calculated VAF represents the proportion of the total effect that is explained by the indirect path and is computed as follows:

𝑎 𝑥 𝑏 + 𝑐′ (1) Where: a × b is the indirect effect, c’ is the direct effect

Based on the interpretation thresholds suggested by Hair et al (2014), a VAF value greater than 80% indicates full mediation, between 20% and 80% indicates partial mediation, and less than 20% suggests no mediation The VAF for the SQ → CS → CL path was 0.847, confirming a full mediation In contrast, the ESP → CS → CL path yielded a VAF of 0.713, confirming a partial mediation effect These results reinforce the central role of customer satisfaction in shaping customer loyalty outcomes within the hospitality context

Table 4.15 Mediation Effect Assessment: Customer Satisfaction as Mediator

As shown in Table 4.15, both mediation paths (H7a and H7b) were supported The findings highlight that improvements in service quality do not directly result in higher customer loyalty unless they enhance the customer's satisfaction first Meanwhile, sustainable environmental practices not only contribute directly to loyalty but also do so more strongly by enhancing customer satisfaction These outcomes emphasize the strategic importance of integrating both service quality enhancement and sustainability initiatives within hotel operations, particularly when aiming to strengthen customer commitment and long-term behavioral outcomes

Discussion of main findings

This study provides new insights into the intricate relationships between service quality (SQ), environmental sustainability practices (ESP), customer satisfaction (CS), customer loyalty (CL), and environmental concern (EC) within the Vietnamese hotel industry While several hypotheses were supported and align with existing hospitality and sustainability literature, others deviated from theoretical expectations, prompting further exploration into the mechanisms underlying these relationships The results reveal both confirmatory and unexpected patterns, reflecting the evolving nature of consumer expectations in service environments that are increasingly shaped by ethical values, sustainability awareness, and affective evaluation

(1) Service Quality does not directly influence Customer Loyalty, but operates fully through Customer Satisfaction

The present study finds that service quality (SQ) exerts a significant and positive effect on customer satisfaction (CS) (H1 supported), consistent with a long-standing consensus in service management and hospitality literature Numerous empirical studies have identified service quality as a multidimensional construct that shapes guests’ evaluative judgments of their hotel experience (Ladhari, 2009b; Parasuraman et al., 1985).Within the hotel sector, dimensions such as Tangibles, responsiveness, reliability, assurance, and empathy have been repeatedly validated as drivers of customer satisfaction (Ali et al., 2016; Wilkins et al., 2007) These findings reaffirm that delivering high-quality service continues to be a fundamental prerequisite for enhancing the affective experience of guests

However, contrary to initial expectations, the direct path from SQ to customer loyalty (CL) was not statistically significant (H3 not supported) This unexpected result challenges earlier research that reported a positive direct link between perceived service quality and customer loyalty in hospitality settings (Caruana, 2002; Kandampully & Suhartanto, 2000) It suggests a paradigm shift in how guests respond to service performance: rather than perceiving quality as a sufficient condition for continued patronage, they may require an additional layer of affective or emotional resonance, represented by satisfaction, before translating evaluations into behavioral loyalty

Supporting this interpretation, the analysis revealed a significant indirect effect of SQ on CL through CS (H7a supported), with a high Variance Accounted For (VAF) of 84.7%, indicating full mediation In other words, service quality only leads to loyalty when it successfully triggers customer satisfaction This pattern is well-aligned with the expectancy-disconfirmation theory by Oliver (1980), where perceived performance alone does not ensure post-purchase behavior unless it surpasses expectations and generates emotional gratification

This result is also consistent with findings by Chi & Gursoy (2009), who showed that satisfaction acts as a mediating mechanism between service quality and loyalty in the hotel industry, particularly when guests engage in affective rather than purely cognitive evaluations Similarly, Pan (2024) highlights that the loyalty-building process in the hospitality industry is increasingly shaped by emotion-laden constructs such as trust, satisfaction, and brand attachment, rather than technical service evaluations alone This shift reflects changing guest expectations, where emotional connection and value alignment often matter more than basic performance standards

In this study of Vietnamese hotel guests with varying environmental awareness, the full mediation effect indicates rising customer expectations As service quality becomes a standard expectation in modern, upscale, or international hotels, it no longer serves as a key differentiator Instead, service quality must deliver satisfying, personalized, and meaningful experiences to foster loyalty behaviors like repeat bookings and positive word-of-mouth.

Recent hospitality research emphasizes the shift toward experience-centric and value-driven guest management, highlighting the importance of integrating functional services with emotional and ethical aspects to foster long-term loyalty (Acampora et al., 2022; Merli et al., 2019) Customer satisfaction serves as a key intermediary, transforming perceptions of service quality into genuine loyalty outcomes that benefit hospitality providers over time.

(2) Environmental Sustainability Practices exert both Direct and Indirect Effects on Loyalty

In contrast to service quality, which affects loyalty only through the mediating role of customer satisfaction, environmental sustainability practices (ESP) exhibit a dual pathway of influence The analysis confirms that ESP significantly enhances customer satisfaction (H2 supported) and also has a marginally significant direct effect on customer loyalty (H4 marginally supported) Moreover, satisfaction was found to partially mediate the ESP–CL relationship (H7b supported), with a Variance Accounted For (VAF) of 71.3%, indicating that while satisfaction plays a substantial role, ESP also contributes to loyalty beyond the experiential level

These findings align with a growing body of literature suggesting that green practices in hotels impact guest perceptions and behaviors through both functional and ethical mechanisms For instance, Berezan et al (2013) found that sustainable hotel practices positively affect guest satisfaction and intention to return Kassinis & Soteriou (2003) also confirmed that environmental management practices contribute to improved customer evaluations and repeat patronage Gao & Mattila (2014) further showed that environmentally proactive actions shape the perception of hotel brands as “modern,”

“progressive,” and “caring,” enhancing affective evaluations Moise et al (2018) added that green practices directly increase both satisfaction and loyalty, while also indirectly influencing loyalty via improved satisfaction levels

More importantly, the marginally significant direct path from ESP to customer loyalty suggests that sustainability efforts also operate as value-expressive signals, helping guests to identify with the hotel’s ethical positioning This observation supports the idea of “moral congruence” between customer values and corporate identity, as theorized in previous research (Han & Kim, 2010; Martínez & Rodríguez Del Bosque, 2013) In other words, even in the absence of a fully satisfying experience, green attributes may independently foster loyalty among guests who prioritize environmental responsibility as part of their consumption identity

Merli et al (2019) reinforce this dual-pathway explanation by showing that staying in a green hotel positively affects revisit intention and word-of-mouth behavior through both satisfaction and direct perceived alignment with environmental values Their empirical evidence demonstrated that guests do not require flawless service to develop loyalty, they also seek authenticity and credibility in sustainability communication This idea is further supported by Han et al (2011), who confirmed that green actions contribute to loyalty by evoking a sense of doing good, not just receiving good service

Taken together, these findings highlight the strategic significance of ESP as both an experiential enhancer and a symbolic differentiator in shaping loyalty In markets like Vietnam, where environmental awareness is rapidly evolving, the presence of visible, transparent, and consistent green practices may increasingly influence not just how customers feel, but who they want to be associated with Thus, hospitality providers should consider ESP as a dual-lever tool: one that strengthens satisfaction while simultaneously appealing to deeper identity-based motivations for loyalty

(3) The Negative Interaction between Environmental Sustainability Practices and Service Quality on Satisfaction

One of the most theoretically intriguing findings of this study lies in the negative interaction effect between environmental sustainability practices (ESP) and service quality (SQ) on customer satisfaction (CS) (H5 marginally supported; β = –0.087, t 1.743, p = 0.081) While both ESP and SQ were individually found to positively influence satisfaction, their combined presence did not result in an additive or synergistic effect Rather, the interaction coefficient suggests that when one factor is already perceived as strong, the marginal contribution of the other to customer satisfaction diminishes, revealing a compensatory or substitution effect

This finding challenges the widely held assumption that combining multiple positive service features always enhances the customer experience Several theoretical explanations help illuminate this unexpected interaction:

First, a substitution or compensatory mechanism may be at work

Customers tend to downplay the importance of either Service Quality (SQ) or Environmental Sustainability Practices (ESP) when they perceive one domain as highly satisfactory; for example, excellent hotel responsiveness and reliability may reduce their attention to environmental efforts Conversely, when ESP such as eco-friendly materials or water conservation becomes salient, customers may shift focus from service performance to value alignment, diminishing the perceived importance of service quality This aligns with research indicating that service features can act as psychological substitutes, especially when customers prioritize ethical and emotional aspects over functional attributes (Gao & Mattila, 2014; Sirdeshmukh et al., 2002).

Second, the result may stem from information overload or value signal saturation

When both ESP and SQ are strongly emphasized, through staff communication, visual cues, certifications, or digital media, customers may encounter cognitive saturation, where the presence of too many strong value signals weakens the emotional and attentional salience of each (Eppler & Mengis, 2004) Prior studies have warned that over-promoting sustainability, especially in conjunction with high service standards, can backfire if customers perceive the communication to be overly assertive, inauthentic, or overwhelming (Gao & Mattila, 2014) This phenomenon may dampen overall satisfaction despite positive intentions

Third, the negative interaction may reflect an expectation–performance gap, particularly under dual excellence

Implications of the Findings

This study offers important contributions to the theoretical advancement of sustainable service research by applying and extending the Stimulus–Organism–Response (S-O-R) framework by Mehrabian & Russell (1974) within the context of the Vietnamese hotel industry In this model, service quality (SQ) and environmental sustainability practices (ESP) operate as external stimuli (S) that shape customer perceptions; customer satisfaction (CS) represents the internal organism (O) - the emotional and cognitive interpretation of these stimuli; and customer loyalty (CL) emerges as the behavioral response (R) The study demonstrates that both functional and ethical attributes of service delivery significantly affect customer satisfaction, which in turn drives loyalty, thereby affirming the relevance of the S-O-R model in hospitality contexts increasingly shaped by sustainability expectations

The findings particularly enrich the theoretical discourse by showing that ESP exerts a dual-pathway influence on customer loyalty, both directly and indirectly through satisfaction, while SQ only affects loyalty indirectly This divergence challenges traditional service evaluation models that privilege performance-based constructs and highlights the symbolic and identity-based role of sustainability in shaping customer behavior Moreover, the study confirms that CS fully mediates the relationship between

SQ and CL, but only partially mediates the ESP-CL link, offering empirical support for the centrality of affective processing in the service-to-loyalty pathway, as emphasized in earlier S-O-R literature (Chi & Gursoy, 2009; Oliver, 1999a)

A further theoretical contribution lies in the nuanced operationalization of environmental concern (EC) as a moderating variable The study finds that EC significantly moderates three relationships in the model: negatively weakening the ESP–

CL and SQ–CS links, while positively amplifying the CS–CL pathway These findings validate recent expansions of the S-O-R framework that incorporate personal values as cognitive filters and emotional amplifiers in service appraisal (Han & Hyun, 2019; Hartmann & Apaolaza-Ibỏủez, 2012) The study also contributes to the green consumer behavior literature by illustrating how value congruence and authenticity shape customer judgments in complex service environments

Finally, the marginally significant negative interaction effect between SQ and ESP on

CS reveals a novel theoretical insight: the combined presence of high performance and strong ethical cues may not always produce additive effects Instead, guests may experience cognitive overload or value substitution, leading to diminishing marginal returns (Eppler & Mengis, 2004; Gao & Mattila, 2014) This observation suggests that future theoretical models should account for nonlinear effects and potential trade-offs when multiple value propositions are emphasized simultaneously Overall, this research affirms the flexibility of the S-O-R model while advancing its application in sustainability-oriented service contexts, encouraging further inquiry into the affective and value-based mechanisms underlying customer loyalty formation

The findings of this study offer several strategic and managerial implications for stakeholders in the hospitality industry, particularly hotel managers, service designers, sustainability officers, and tourism policymakers, who are navigating the growing demand for both high-quality service and environmental responsibility in an increasingly competitive and value-driven market like Vietnam

First, while service quality (SQ) remains foundational to guest satisfaction, its effect on loyalty is not automatic The analysis showed that customers only become loyal when their service expectations are not merely met, but emotionally fulfilled through satisfaction For hotel managers, this means that improving technical service dimensions

- such as responsiveness, assurance, and empathy - should be accompanied by consistent efforts to create positive emotional experiences Training programs, frontline empowerment, and a culture of personalized care remain essential investments

Second, environmental sustainability practices (ESP) have emerged not only as experience enhancers but also as value signals that can independently foster customer loyalty Even moderately visible actions, such as in-room recycling options, visible energy-saving systems, or the use of sustainable materials, can contribute to brand attachment, especially when communicated transparently Hotels that integrate sustainability meaningfully into their operations are more likely to resonate with value- driven customers who seek alignment between their consumption choices and ethical beliefs

Customer satisfaction plays a vital mediating role between service quality, green practices, and guest loyalty, highlighting the need for hotels to focus on both operational excellence and emotional connection Loyalty strategies should go beyond service improvements to foster trust, shared values, and personal rewards, leveraging storytelling, staff engagement, and green initiatives to boost satisfaction Enhancing these emotional and trust-based elements can significantly strengthen guest loyalty and reinforce sustainable practices.

Importantly, the moderating role of environmental concern (EC) suggests the need for nuanced segmentation Guests with higher ecological awareness may require more than symbolic gestures, they respond to authenticity, consistency, and genuine impact For these segments, hotels should go beyond surface-level initiatives and offer deeper engagement opportunities, such as carbon offset packages, locally sourced amenities, or behind-the-scenes sustainability tours Communication strategies should emphasize not just what the hotel is doing, but why it matters and how guests can be part of the solution

Finally, the negative interaction found between service quality and ESP on satisfaction serves as a caution: more is not always better When both domains are overemphasized without a coherent narrative, guests may feel overwhelmed or confused Instead of maximizing all dimensions simultaneously, managers should calibrate their messaging and delivery to suit specific customer profiles, context, and stay purpose Clear storytelling- framing the hotel’s identity around a unifying value, whether it be comfort, ethics, or innovation, may help avoid dilution of impact and create a stronger emotional bond

In summary, this study affirms that environmental sustainability and service quality are not competing priorities but complementary pillars of long-term customer engagement Hospitality businesses that can harmonize operational excellence with authentic ethical positioning will not only earn guest loyalty but also contribute meaningfully to broader social and environmental goals.

CONCLUSIONS