Understanding the effect of millennial tourists’ accommodation service experiences using structural equation modeling techniques: an emerging economy context

Abstract

Purpose

The purpose of this study is to investigate the role of millennial tourists’ accommodation service experiences (ASEs) on their satisfaction, word-of-mouth (WOM) and revisit intentions (RIs) in an emerging economy.

Design/methodology/approach

A survey instrument was used to collect cross-sectional data from millennial tourists, and the 282 valid datasets were analyzed using structural equation modeling.

Findings

The results demonstrated that ASE had a beneficial impact on satisfaction and WOM but not on RI. Significant positive associations between tourist satisfaction, WOM and RI were also discovered. Additionally, WOM research sheds new light on how the ASE of millennial tourists affects their satisfaction, WOM and propensity to return. Furthermore, results show that WOM intentions and satisfaction mediate the relationship between ASE and RIs.

Originality/value

The study presents a unique research context, the application of advanced statistical techniques and the comprehensive investigation of key outcome variables in the context of millennial tourists’ ASEs in an emerging economy. This study contributes significantly to the body of knowledge in the field of tourism research, aiming to meet long-term goals in a sustainable way for the hospitality industry operators by integrating ASE, satisfaction, WOM and RI. Additionally, the study presents the mediating role of WOM and satisfaction in the millennial tourists’ emerging country context.

Citation

Hossain, M.S., Hossain, M.A., Masud, A.A. and Hossain, M.S. (2024), "Understanding the effect of millennial tourists’ accommodation service experiences using structural equation modeling techniques: an emerging economy context", South Asian Journal of Marketing, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/SAJM-02-2023-0019

1. Introduction

Globally, accommodation services play an eminent role in the development of the tourism sector; Deng et al. (2013) mentioned that accommodation is one of the most important components of the tourism industry. Travelers are eager to experience the tastes of the local cuisine and accommodation services at their destination (Nekmahmud and Hassan, 2021) for their happiness. Tourists’ happiness is directly related to service quality, and there is a clear, positive link between customer satisfaction and word-of-mouth (WOM) (Gholipour Soleimani and Einolahzadeh, 2018). The quality of a visitor’s accommodation service experience (ASE) has an impact on the duration of their stay, the activities they undertake and the way they spend their time when visiting a tourist destination (Rauch et al., 2015). Different visitors have diverse experiences, such as millennial tourists who were born between 1980 and 1999 or between 1980 and 2000 (Visit Scotland, 2017). They consider travel to be a top priority in their lives, and believe that travel and tourism are indeed very crucial and necessary (Cavagnaro et al., 2018), which has had a great impact on the world’s tourism service industry as well as the context of emerging countries.

As an emerging country, Bangladesh’s domestic tourists, especially millennials who make up 38% of the population of the country (Ghosh, 2019), play an important role in the tourism industry. Bangladesh is now experiencing a “demographic dividend” period (Hayes and Jones, 2015) with a large number of millennial tourists who are traveling a lot (Halder and Sarker, 2022), but there is still a dearth of research on this group. This study focused on the millennial generation, and millennials’ ASE and satisfaction, WOM and revisit intentions (RIs). Millennials’ behavior as green customers (Naderi and Van Steenburg, 2018) and their willingness to spend on green restaurants (Naderi and Van Steenburg, 2018) and their satisfaction toward the hotel industry in Malaysia (Shafiq et al., 2019) have been studied. But in the context of millennials’ ASE, satisfaction, WOM and return intentions, there are still areas of study to be done on this target group in an emerging country context (Hossain et al., 2023). Being an emerging nation, Bangladesh will mostly see a rise in tourism from young travelers (Parvez and Jahid Bin Kashem, 2018) and, thus, the selection of this study topic is timely and appropriate.

To our knowledge, limited studies have yet been created that constitute the core of the millennial tourist’s ASE in an emerging economic tourism setting. In light of the literature’s existing research gaps, this study offers a new perspective on ASE using a multidimensional and hierarchical model that includes accommodation infrastructure, room facilities, safety and security, and staff attitude and behavior dimensions for the millennial generation. Accordingly, the following are our research questions (RQ):

RQ1.

Does domestic millennial tourists’ ASE improve satisfaction, WOM and RI in an emerging country context?

RQ2.

Does domestic millennial tourists’ WOM intentions and satisfaction mediate the relationship between ASE and RIs in an emerging country context?

Since there is not much study on the subject, analyzing the connections between these constructions will provide future researchers, tourist professionals and business executives a more complete knowledge. First, by examining a multidimensional aspect of lodging services as a first research will make it simpler for hotel managers and decision-makers to comprehend how consumers choose to employ lodging services in the tourism business. Second, the findings of this study will demonstrate the amazing influence of ASE aspects on behavioral perceptions, offer significant light on the likelihood of repeat visits and make important contributions to the understanding of visitor satisfaction. Third, the findings will subsidize existing literature by conducting an examination into a number of service marketing ideas, which will increase our comprehension of tourists’ perceptions of WOM and satisfaction. Fourth, our study findings will recommend to managers and stakeholders that they be able to capture this millennial group’s market share by using strategic experience design, which will enhance the sustainability of the hospitality industry. Finally, our findings may greatly help tourism professionals and hotel managers in creating and maintaining service marketing strategies that improve the quality of their services, increase the satisfaction of long-term visitors and thus encourage them to return.

2. Review of relevant literature and hypothesis development

2.1 Millennial tourist and accommodation service experiences

Millennial tourists have recently emerged as a niche market in the hospitality industry, bringing with them new perspectives and attitudes and posing unique customer experience challenges (Mhlanga, 2018). Enlightenment on how to create memorable travel experiences is critical for the tourism industry because these experiences influence future choices, loyalty and WOM intention, particularly among the millennial generation, which accounts for 23% (1.8 billion) of the universal population (Neufeld, 2021). Consumer experience is frequently portrayed as a multidimensional, holistic construct capturing customer reactions to corporate or brand interactions (Behl et al., 2022). In the hospitality industry, satisfaction with hotels showed that cooperation between hotel guests and service workers was important (Clemes et al., 2010), and tourist experiences were studied using different dimensions.

In a recent study, Chittiprolu et al. (2021) found that tourists in a developing nation environment complained about hotels’ intangible service difficulties, including staff attitude, service failure, problems with reservations and meals, value for money and room condition. Additionally, Nunkoo et al. (2019) looked at the infrastructure of the accommodation, the attitude and behavior of the staff, the interaction between customers, the expertise of the staff, the quality of the food and drinks, the quality of the front desk, the quality of the room, the safety and security, the sociability and the amount of time spent waiting to understand customer satisfaction in terms of tourists’ ASE.

In the hospitality and tourism industries, an experience can be anything a tourist does or sees, thinks or feels, says or doesn’t say, or is expressed or implied and accommodation infrastructure, room facilities, safety and security, attitude and behavior of staff are all sub-dimensions of the ASE (Marôco and Marôco, 2013; Nunkoo et al., 2019). In this study, we used four dimensions: “accommodation infrastructure,” “room facilities,” “staff attitude and behavior,” and “safety and security” to look at how the experience of staying in accommodation affects things like millennium tourists’ satisfaction, WOM and intentions to return (RI).

The quality of the hotel service experience influences tourist satisfaction and behavioral intentions (e.g. return and WOM) (Marôco and Marôco, 2013; Nunkoo et al., 2019; Hossain et al., 2021). Additionally, tourists who are pleased with their accommodations and dining options are more inclined to return and recommend the place (Mai et al., 2019). Tourists are more inclined to return to a location if they have a comfortable experience with lodging and a portion of delicious food (Mai et al., 2019). The experiences of visitors enhance their satisfaction, advocacy and perception (Hossain et al., 2022). In the study context of an emerging market perspective, guest experiences in accommodations have a favorable and considerable impact on visitors’ satisfaction and propensity to return (Ugwuanyi et al., 2021).

Therefore, we hypothesize that the ASE of millennials in an emerging economy:

H1.

Millennials’ ASE has a positive effect on their satisfaction.

H2.

Millennials’ ASE has a positive effect on their WOM intentions.

H3.

Millennials’ ASE has a positive effect on their RIs.

2.2 Millennial tourists’ revisit intentions and positive WOM

Past studies have shown the relevance of repeat tourists for the long-term sustainability of any economy’s tourism industry (Stylos et al., 2017). A study by Um et al. (2006) found that relying on first-time visitors is more expensive than relying on repeat visitors. This means that the tourism sector of a location’s economy is very dependent on repeat visitors. Given the significance of repeat tourism, previous studies have made significant efforts to understand the antecedents of visitors’ return intentions, and it has been discovered that satisfaction with the venue is a key driver, in accordance with results from consumer behavior research. The relevance of visitors’ experiences in affecting their decision to return was underlined by Zhang et al. (2017). Similarly, numerous previous empirical studies have shown that visitors’ experiences and contentment with a venue are important drivers of their inclination to return (Ranjbarian and Pool, 2015; Um et al., 2006).

As such, we formulate three additional hypotheses in line with our study purpose:

H4.

Millennial tourists’ satisfaction generates positive WOM intentions.

H5.

Millennial tourists’ satisfaction accelerates their RIs.

H6.

Millennial tourists’ positive WOM revives accommodation RIs.

Despite the fact that such research has a high potential to contribute to the related literature, there are few scholarly investigations on millennial tourists’ perceptions of ASE in emerging economies, which evidently lack the hospitality facilities of tourist places, and their contribution to overall satisfaction and behavioral intentions (WOM and RIs). Delightful experiences are commonly accompanied by satisfaction, and visitors who have had a positive experience are more likely to return (Kim, 2018). Iqbal et al. (2017) have focused on the role of several mediating elements in the relationship between visitor experiences and return intention. Customer happiness, according to Iqbal et al. (2017), plays a partly mediating role in the link between service quality and customer loyalty. Tourists’ satisfaction (Binti Mohd Ali et al., 2020) and WOM also worked as a go-between for hotel clients’ readiness to purchase and the level of service they received (Kasa et al., 2018).

Therefore, we hypothesize that the ASE of millennials in an emerging economy:

H7.

Millennials’ WOM intentions mediate the relationship between ASE and RIs.

H8.

Millennials’ satisfaction mediates the relationship between ASE and RIs.

3. Research methodology

3.1 Sample design and study context

The research has adopted a quantitative approach to tourist psychological study, including quantitative data collection and analysis methods. In the beginning, we thoroughly reviewed the body of literature that already exists in our study area, noting key points, significant variables and their relationships, and then we drafted a preliminary study plan. Thereafter, we consult with a focus group that consists of two people from the tourism managers, two tourist scientists and one hotel owner. Later, we scheduled a meeting with young tourists to learn more about their perspectives on tourist destinations, especially in a constructive way. We agreed to work on the suggested study model in the Bangladesh context which is depicted in Figure 1 after considering the proposals and recommendations from all three layers.

A structured survey questionnaire was developed, and offline and online surveys were conducted using the Google platform to collect cross-sectional data. The data were gathered using a convenience sampling approach. The questions of the survey were developed keeping in mind the objectives of the research. The survey investigates how young travelers now feel about their accommodation experiences, as well as how satisfied they are and whether they plan to return. A standardized questionnaire with a Likert scale ranging from strongly disagree (“1”) to highly agree (“7”) was used to assess the components. Where necessary, adjustments and alterations were made. Online and offline surveys were performed over a three-week period in March 2021. Before the main survey, a pilot test was carried out to view the response pattern. After accommodating the minor modifications and changes, we finalized the data collection instruments.

This research is conclusive because hypotheses are established to validate the relationship between variables. According to the study’s objective, we chose the hierarchical model of the tourism industry in Bangladesh due to the country’s rising and densely populated characteristics and the emerging country context in the world. Domestic tourists, particularly millennials, who make up 38% of Bangladesh’s population, play a major role in the country’s tourism business (Ghosh, 2019). Although the tourist service sector is highly competitive and contributes significantly to the GDP, there are few studies on the sector and no quality measurement standards.

3.2 Sample and data

This study surveyed Bangladeshi tourists to test their behavioral attitudes and hypothesized associations. The participants were encouraged to submit valuable comments to help the investigation. In total, 541 questionnaires were collected using the convenience sampling technique, and after data cleaning and multivariate normality checking, only 282 samples were kept (Jahan and Shahria, 2021).

Table 1 summarizes the sample statistics. The majority of participants were male (56.4%), and 59.6% of them had a higher school degree. Overall, 93.3% of them are single. The age distribution of the participants is between 16 and 25 years (67%), 26 and 35 years (30%), and only 3% are over 35 years old, which shows that respondents are millennial tourists.

3.3 Instrument design

The study’s measurement tools were adapted from previously validated measures or derived from prior research. ASE measures (e.g. accommodation infrastructure, room facilities, staff attitude and behavior, safety and security) are adapted from the research of Sangpikul (2018), Nunkoo et al. (2019), Marôco and Marôco (2013), and Mai et al. (2019). Tourist satisfaction is adapted from the research of Khadaroo and Seetanah (2008), and RIs and WOM intentions are based on the works of Stylos et al. (2017), Um et al. (2006), and Ranjbarian and Pool (2015). When all elements belong to the same concept or the first factor explains the greatest variation, Common Method Variance (CMV) problems occur. The unrotated solution shows that the first component accounted for 41.47% of the variance and that other components had eigenvalues higher than 1. This shows that the data did not have a CMV problem (Podsakoff et al., 2003).

4. Empirical results

With the statistical program Amos-24, which is often used to test and prove theories, confirmatory factor analysis and structural equation modeling were done on the proposed research model. In terms of parameter consistency and accuracy, this method exceeds partial least squares regression. It helps a lot to prove that the measurement model is accurate and reliable, and it shows how the variables in the structural model are connected theoretically (Fornell and Larcker, 1981). In addition, this study performs bootstrapping analyses to assess the mediating effects of the proposed model.

In order to check the validity and reliability of the data, we first validated the measurement model analysis (e.g. convergent and discriminant validity, model fit test, etc.). After receiving good feedback on the validity and dependability of the data, we moved on to testing the hypothesis in a structural model, which will actually provide a response to RQ1. Thereafter, we examined mediation effects in our model, which certainly addressed the RQ2.

4.1 Measurement model’s analysis

Two distinct measurement models were created in this investigation. Convergent and discriminant validity are evaluated to ensure the measurement models’ validity and reliability. The following sections discuss the fitness, validity and reliability of the measurement models (e.g. the first-order and higher-order models). As seen in Table 2 and Figure 2 of the first-order measurement model, all measures are robust in terms of reliability. For all ASE sub-dimensions, all standardized factor loadings are statistically significant and higher than their minimal criterion of 0.70 (Fornell and Larcker, 1981). All constructions have Cronbach’s alpha values above the minimal critical value of 0.70 (Hair et al., 2010). The proposed critical threshold of 0.70 is exceeded by the composite reliability (CR), which varies from 0.88 to 0.93 (Hair et al., 2010). The average variance extracted (AVE) for each sub-dimension is >0.50, with values in the range of 0.65–0.80, appropriate for the model fit. Additionally, all sub-dimensional component pair correlation coefficients are lower than the lowest square root of AVE values (0.80–0.89). In the first-order model, the maximum correlation value is 0.82, suggesting good discriminant validity (Hair et al., 2010). As a result, both convergent and discriminant validity are satisfied.

As shown in Table 3, all measures are robust, with standardized regression weights whose accommodation infrastructure values range from 0.77 to 0.85, and the t-value is a suggested cut-off value (Hair et al., 2010). As the same room facilities, staff attitude-behavior, safety and security, tourists’ satisfaction, WOM intention and RI have values ranging from 0.70 to 0.88, 0.78 to 0.88, 0.88 to 0.91, 0.83 to 0.86, 0.82 to 0.88 and 0.85 to 0.92, and the t-value is a suggested cut-off value (Hair et al., 2010).

The measurement model’s overall fit statistics are found to meet their respective cut-off criteria, including the following values: Tucker–Lewis index (TLI) = 0.972, goodness-of-fit index (GFI) = 0.88, adjusted GFI (AGFI) = 0.84, comparative fit index (CFI) = 0.976 and root mean square error of approximation (RMSEA) = 0.042 (Hu and Bentler, 1999). As a result, the model’s fit is excellent. Following that, one new model (see Figure 3) is created using a second-order reflective construct known as experience. Accommodation infrastructure, room facilities, staff attitude, behavior, safety and security are the sub-dimensions of experience. These sub-dimensions are analyzed based on the significance of their corresponding factor loads on experience. However, a higher-order measurement model is tested using one second-order reflective construct and four additional psychometric variables. The findings indicate that the reliability, divergent and discriminant validity, multicollinearity and model fit indices are satisfactorily archived (see Table 4).

Finally, all parameters in the higher-order measurement model meet their particular cut-off criteria, including Chi-square = 544.721 with 357 degrees of freedom (χ²/d = 1.52), GFI = 0.878, AGFI = 0.851, CFI = 0.974, TLI = 0.970 and RMSEA = 0.043 (Hu and Bentler, 1999). As a result, the model’s fit is excellent (see Table 4).

4.2 Structural model analysis

Based on the fitness of the measurement model, we proceeded to conduct Structural equation modeling (SEM) (Figure 4) aimed at assessing the hypothesized paths. Table 5 shows that the SEM model has an adequate model fit to the data (X2/d = 1.52, GFI = 0.83, AGFI = 0.851, CFI = 0.974, TLI = 0.970, Incremental Fit Index (IFI) = 0.94, Normed Fit Index (NFI) = 0.87, RMSEA = 0.043). The data show that the model explained (e.g. R2 value) 62%, 73% and 51% of the variance in WOM intention, RI and satisfaction, respectively. Model fit indices are in their critical ranges for the overall model (Bagozzi and Yi, 2011).

Out of six hypotheses (direct effect), all are statistically supported at p < 0.001, except for hypothesis number 3, ASE and RI (ᵦ = −0.00, t = −0.11, p < 0.001). The result shown in the figure revealed positive and significant relationships between ASE to satisfaction and WOM (ᵦ = 0.71, t = 9.98, p < 0.001), (ᵦ = 0.36, t = 4.58, p < 0.001); thus, H1 and H2 are accepted. Again, the result shows a positive and significant relationship between satisfaction with WOM intention and RI (ᵦ = 0.48 t = 6.16, p < 0.001), (ᵦ = 0.32, t = 4.38, p < 0.001); thus, H4 and H5 are accepted. Lastly, the result shown in the figure revealed a positive and significant relationship between WOM intention and RI (ᵦ = 0.59, t = −7.69, p < 0.001), so the H6 is accepted. It is discovered that the values are within the acceptable range (Hu and Bentler, 1999). Therefore, we can draw the conclusion that our RQ1 - Does domestic millennial tourists’ ASE improve satisfaction, WOM and RI? - is satisfactorily addressed.

4.3 Mediation model’s analysis

This study examines the role of WOM in mediating ASE and RI, as well as the role of tourist satisfaction in mediating the ASE and RI. Partial mediation occurs when the dependent and independent variables have a statistically significant connection with the mediator, and complete mediation occurs when the dependent and independent variables have a statistically significant link through the mediator (Baron and Kenny, 1986).

In total, 282 samples were used for bootstrapping analysis, and 5,000 cycles were made with 95% confidence intervals (Table 6). WOM and tourist satisfaction mediate this study’s ASE and subsequent RIs. Therefore, hypotheses H7 and H8 are supported by significant partial mediation effects, which certainly answered the yes; RQ2 - Does domestic millennial tourists’ WOM intentions and satisfaction mediate the relationship between ASE and RIs?

5. Discussion

The research investigates the satisfaction, WOM impact and intention to revisit a destination among millennial tourists using a multidimensional ASE framework. It shows how WOM and satisfaction act as a bridge between ASE and wanting to go back to the same place. Our conceptual model substantiated that ASE is a multidimensional object, revealing accommodation infrastructure, room facilities, staff attitude and behavior, and safety and security as the prime factors in building experiences around the accommodation. The structural model shows good explanatory power in predicting 62% of the variance in WOM, 51% of the variance in millennial tourist satisfaction and 73% of the variance in RI. Furthermore, the data analysis results show that most of the hypothetical relationships are supported by a good model fit.

Precisely, in accordance with our proposed theoretical base on ASE, the results of the study indicate that accommodation infrastructure, room facilities, staff attitude and behavior, and safety and security are the prime building blocks of ASE. These results are consistent with previous studies by Mai et al. (2019), Nunkoo et al. (2019), and Marôco and Marôco (2013), who considered accommodation infrastructure, room quality, safety and security, attitude and behavior of employees, and customer interaction as the key dimensions of ASE and showed that these have a significant influence on tourist satisfaction and loyalty.

Hence, ASEs are not a unidimensional element but rather a multidimensional formation with several diversified experiences that significantly engender positive perceptions toward repeating truisms among millennial tourists. Accordingly, the results from the SEM analysis prove that ASE has a significant and positive influence on WOM, which supports the earlier study findings of Nunkoo et al. (2019) and Marôco and Marôco (2013). They reported that hotel service quality, based on the dimensions of the behavior of staff, reception, safety and security, room facilities, bar and restaurant, has a notable impact on tourist satisfaction and loyalty. Consequently, it is evident that the physical condition of a hotel or motel, proper environmental protection, equitable utilities and the ambience of a room, as well as the friendly interaction of employees, have been more emphasized by young tourists when staying in a tourist spot.

Nevertheless, this study didn’t find a significant influence of ASE on RI. This absence of significance may be attributed to the relatively lower priority accorded to it for subsequent visits by young tourists. Additionally, it is plausible that our target participants are young tourists who have a greater tendency to discover something new, and thus they may not consider the same hotel or motel based on their previous experience. Consequently, in catering to the needs of this demographic, ASE factors such as employee behavior, reception, safety and security, hotel amenities, bars and restaurants should be assigned heightened importance. This strategic emphasis is essential for cultivating a contented and loyal tourist clientele.

Moreover, tourist satisfaction was found to have a positive and significant influence on both WOM and RI, aligning with the findings of Deng et al. (2013), and Marôco and Marôco (2013). When tourists experience convenient, equitable and safe accommodation, this satisfaction can inspire them to recommend the establishment to their peers, colleagues, neighbors, friends and family members, further increasing the likelihood of repeat visits. A higher level of tourist satisfaction also leads to increased trust and reduced uncertainty, ultimately resulting in more frequent visits and referrals.

Furthermore, our results reinforce the positive association between WOM and RI, as evidenced by previous studies (Kim, 2018; Hossain et al., 2021, 2023). This suggests that memorable tourism experiences that evoke excitement, enjoyment and rejuvenation prompt tourists to share their positive experiences with others, effectively creating organic marketing. Given the cost-effectiveness, simplicity and dynamic nature of WOM, it should be prioritized in marketing strategies, particularly within the tourism industry.

Furthermore, our study reveals significant mediating effects of tourist satisfaction and WOM on the relationship between ASE and RI. Interestingly, while no direct effect of ASE on RI was found, our results demonstrate significant indirect effects achieved through WOM and satisfaction. This outcome further underscores the critical role of tourist recommendations and their overall satisfaction. As RI is considered a potent weapon in the competitive landscape, focusing solely on ASE is insufficient. To effectively encourage repeat visits, prioritizing customer satisfaction and organically generated WOM is essential (Hossain et al., 2022).

Therefore, accommodation dimensions such as room facilities, safety and security, staff behavior and accommodation infrastructure are not only precious to increasing tourist satisfaction and producing WOM (Hossain et al., 2023) but also have a huge strengthening power to encourage RI with the intervening factors of satisfaction and WOM. In other words, where there is a higher degree of tourist satisfaction and WOM, there is a greater direct influence on RI and that provokes strengthening the association between ASE and RI. As a result, tourists will be less likely to object to paying more for future visits if they have had refreshed, enjoyable and equitable lodging experiences. Based on the foregoing, we believe that the findings of our study will strengthen theoretical knowledge on young tourists’ ASE and their effects on tourist satisfaction, WOM and RI with multiple mediating relationships, as well as serve as a key underpinning research framework for future research. Overall, we can say that our two research questions have been satisfactorily answered, and it makes sense that domestic millennial visitors’ ASE will increase satisfaction, WOM and intention to return. We have also shown that the association between ASE and RIs is mediated by the WOM intents and satisfaction of domestic millennial tourists.

Beyond theoretical implications, our study offers a number of business implications for fostering long-term customer loyalty in the hospitality industry.

First, our findings showed that accommodation infrastructure, room facilities, staff behavior, and safety and security should be prioritized in building a pleasant atmosphere for accommodation services. Accommodation service providers should take affirmative action in designing and decorating accommodation layouts, interiors, parking, other facilities, etc. The room should be well equipped with aesthetics and sophisticated appeal. For instance, in order to ensure safety and security, practitioners should install techno-human systems that will build tourists’ confidence to stay longer. Readily available fire extinguishers further bolster safety. Practitioners should hold staff training programs to foster courtesy, professionalism and commitment to guest satisfaction and security.

Second, fostering positive interactions with guests is crucial for boosting satisfaction and repeat visits. For instance, the manager should deliberately talk with the tourists, organize social gatherings and address service-related concerns diligently. Maintaining a record of incidents and implementing a customer complaint system facilitates continuous improvement. In these situations, the management may set up some additional rewards to encourage employee involvement in the task to lower customer complaints or may impose some sanctions for this (Hossain et al., 2022).

Third, minimizing waiting times and personalizing services can transform existing guests into brand advocates. In these regards, tourist-specific demands such as special arrangements of tourist wishes, arranging pick-up and drop-off services from station-accommodation-tourist spots, etc. can be entertained, and thereby positive referrals are inevitable.

Fourth, acknowledging the direct and mediating effects of tourist satisfaction and WOM emphasizes the importance of offering value-added services. In this regard, managers may incorporate virtual tourism to attract senior citizens, highly busy tourists and physically challenged tourists. By integrating artificial intelligence, virtual live tourism will be a new discovery for many, especially young tourists. Finally, because RI has become increasingly important in the context of fierce competition among hotel and motel service providers, practitioners should employ the mechanism demonstrated in the current study to establish a long-term tourist base.

6. Conclusion

Our study identified ASE dimensions and investigated their influences on tourist satisfaction, WOM and RI, along with multiple mediating effects. By developing a conceptual framework, we address two research questions: RQ1: Does domestic millennial tourists’ accommodation experience improve satisfaction, WOM and RI? RQ2: Do domestic millennial tourists’ WOM intentions and satisfaction mediate the relationship between ASE and RIs?

The findings confirmed that overall ASE is a reflection of accommodation infrastructure, staff behavior, room facilities, and safety and security, thereby positively affecting tourist satisfaction and WOM. Furthermore, tourist satisfaction is an important variable for WOM and RI, and WOM for RI. Notably, our study demonstrates that WOM and tourist satisfaction act as significant mediators between ASE and the RI of young tourists in the context of Bangladesh, with potential applicability to similar settings.

Despite the significant theoretical understandings and practical implications, the current study has few limitations, as other studies do. First, caution is advised in generalizing the results of the study because it is based only on young tourists aged between 18 and 30. Our study mainly focused on young tourists, so the results may not be the same with any other age group. Therefore, we suggest that future researchers validate our model with a more diversified age group so that it will be more generalized. Second, as this is a survey-based study, there are inevitable problems in responding to the questionnaire. Participants may not be fully free to respond to what they exactly want, or they may not be properly scored. Thus, we expect more longitudinal studies to overcome the survey-based studies. Finally, the current study only considers ASE, tourist satisfaction, WOM and RI, leaving many other important psychometric variables. Future research could extend our model by incorporating such factors for a more comprehensive understanding.

Figures

Figure 1

Proposed research framework

Figure 2

First-order measurement model

Figure 3

Higher-order measurement model

Figure 4

Structural model

Table 1

Demographic characteristics (n = 282)

Variable	Items	Percentage
Gender	Male	56.4%
	Female	43.6%
Age	16–25	67%
	26–35	30%
	+35	3%
Educational qualification	HSC	59.6%
	Graduation	38.7%
	Post-Graduation	1.8%
Marital status	Married	6.7%
	Single	93.3%

Source(s): Survey data

Table 2

First-order construct validity statistics

Constructs	CR	AVE	MSV	MaxR(H)	α	1	2	3	4	5	6	7
1. Staffs attitude-behavior	0.917	0.689	0.461	0.921	0.919	0.830
2. WOM intentions	0.922	0.748	0.682	0.925	0.926	0.649	0.865
3. Room facilities	0.902	0.650	0.585	0.917	0.913	0.679	0.614	0.806
4. Revisit intentions	0.934	0.781	0.682	0.941	0.930	0.629	0.826	0.553	0.884
5. Safety security	0.926	0.807	0.342	0.927	0.925	0.497	0.385	0.584	0.326	0.898
6. Satisfaction	0.882	0.714	0.576	0.883	0.882	0.640	0.744	0.633	0.759	0.367	0.845
7. Accommodation infrastructure	0.912	0.674	0.585	0.914	0.914	0.670	0.587	0.765	0.512	0.585	0.590	0.821

Note(s): Model fit indices: X²/d = 1.500, GFI = 0.884, AGFI = 0.854, CFI = 0.976, TLI = 0.972, IFI = 0.976, NFI = 0.931, RMSEA = 0.042, P-Close = 0.961. Italic diagonal values are the square root of the AVE value

Source(s): Survey data

Table 3

Standardized regression weights

Variable	Items	Estimate	S.E.	t-value
Accommodation infrastructure	ACI1	0.812
	ACI2	0.783	0.052	17.482
	ACI3	0.840	0.066	16.181
	ACI4	0.859	0.065	16.690
	ACI5	0.808	0.066	15.343
Room facilities	RF1	0.772
	RF2	0.772	0.053	19.226
	RF3	0.706	0.078	13.462
	RF4	0.886	0.079	15.921
	RF5	0.880	0.080	15.815
Staffs attitude-behavior	SAB1	0.844
	SAB2	0.883	0.045	23.450
	SAB3	0.838	0.063	16.377
	SAB4	0.781	0.068	14.902
	SAB5	0.800	0.067	14.913
Safety and security	SS1	0.881
	SS2	0.898	0.048	21.264
	SS3	0.915	0.046	21.909
Tourists’ satisfaction	TST3	0.862
	TST4	0.841	0.055	17.245
	TST5	0.832	0.056	16.965
WOM intention	WOM1	0.827
	WOM3	0.862	0.058	17.379
	WOM4	0.886	0.058	18.199
	WOM5	0.883	0.058	18.242
Revisit intention	RI1	0.855
	RI3	0.851	0.055	17.480
	RI4	0.898	0.050	20.079
	RI5	0.929	0.054	18.684

Source(s): Survey results

Table 4

Higher-order construct validity statistics

	CR	AVE	MSV	MaxR(H)	1	2	3	4
1. Satisfaction	0.882	0.714	0.576	0.883	0.845
2. WOM intentions	0.922	0.747	0.682	0.924	0.745	0.865
3. Revisit intentions	0.934	0.781	0.682	0.940	0.759	0.826	0.884
4. Experience	0.875	0.639	0.511	0.896	0.715	0.710	0.643	0.799

Note(s): Model fit indices: X2/d = 1.526, GFI = 0.878, AGFI = 0.851, CFI = 0.974, TLI = 0.970, IFI = 0.974, NFI = 0.928, RMSEA = 0.043, p-close = 0.940. Italic diagonal values are the square root of the AVE value

Source(s): Survey data

Table 5

Hypothesis results

Hypothetical paths	Estimate	S.E.	t-value	p-value	Decision
H1: ASE → WOM Intention	0.364	0.087	4.585	***	Accepted
H2: ASE → Satisfaction	0.715	0.079	9.983	***	Accepted
H3: ASE → Revisit intention	−0.008	0.082	−0.115	0.908	Rejected
H4: Satisfaction → WOM Intention	0.485	0.078	6.161	***	Accepted
H5: Satisfaction → Revisit intention	0.325	0.081	4.388	***	Accepted
H6: WOM Intention → Revisit intention	0.590	0.085	7.696	***	Accepted

Note(s): X²/df. = 1.526, AGFI = 0.851, CFI = 0.974, TLI = 0.970, RMSEA = 0.043, P-close = 0.940

***p < 0.001

Source(s): Survey data

Table 6

Mediation model results

			Bias-corrected
			95% CI
	Indirect effect	Estimate	Lower	Upper	p	Decisions
H7	Experience → WOM intentions → revisit intentions	0.259	0.026	0.714	*	Accepted
H8	Experience → Satisfaction → revisit intentions	0.280	0.067	0.792	**	Accepted

Note(s): ***p < 0.001; **p < 0.01; *p < 0.05, CI = confidence interval, the process repeated 5,000 times

Source(s): Survey data