The Impact of IMC Messages on Gen Z’s Luxury Hotel Booking Intentions: A Case Study of Marriott’s “The Luxury Collection” ()
1. Introduction
1.1. Research Background and Problem Statement
The global luxury travel and high-end hotel market is currently undergoing a structural transformation, primarily driven by generational succession. As digital natives and an increasingly influential consumer cohort, Generation Z (typically defined as those born between 1997 and 2012) (McCrindle Research, 2023) is fundamentally redefining the very essence of luxury consumption. In contrast to the traditional logic of luxury, which centered on material possession, social status signaling, and exclusivity, Gen Z consumers place greater emphasis on experiential value, emotional resonance, and personal meaning. They often view consumption as an act of self-expression and a reflection of their values (Francis & Hoefel, 2018). This is particularly evident in the context of travel and hospitality, where Gen Z shows a marked preference for immersive experiences, authentic local cultural connections, and luxury formats that demonstrate brand authenticity and social responsibility.
Extant research indicates that Gen Z exhibits a distinct value-oriented approach in their brand decision-making. They show significant concern for brands’ performance in sustainability, social responsibility, and ethical stances, expecting transparency and consistency in marketing communications (Halibas et al., 2025). As deep-seated users of social media, their consumer decision journey is highly fragmented, shaped by multi-channel, multi-touchpoint information, which underscores the critical role of Integrated Marketing Communications (IMC) in shaping their brand attitudes and behavioral intentions. However, the prevailing marketing paradigm in the luxury hotel industry, which still heavily relies on emphasizing physical amenities, brand heritage, and scarcity (Giousmpasoglou & Marinakou, 2024), is creating a growing tension with Gen Z’s pursuit of “experiential luxury” and “value-driven luxury”. The industry’s traditional focus on “symbolic luxury” is gradually becoming misaligned with the new generation’s expectations.
Against this backdrop, Marriott International’s The Luxury Collection brand presents a compelling case for investigation. With its core positioning as “a constellation of unique hotels and resorts worldwide, each serving as a gateway to the destination’s soul”, the brand emphasizes destination culture, local narratives, and unique experiences (Hospitality Net, 2024).
Current research on luxury brands and Gen Z has predominantly focused on sectors such as fashion, beauty, and fast-moving consumer goods (Thaichon & Quach, 2023). Empirical studies systematically examining the impact mechanism of IMC message content on the booking and payment intentions of Gen Z consumers within the specific context of the luxury hotel industry remain relatively scarce. This research gap not only limits the academic understanding of the communication logic behind “new luxury consumption” but also leaves hotel practitioners with insufficient evidence-based guidance for crafting effective marketing strategies targeting the younger generation.
Based on the above theoretical and practical background, this study addresses the following core research question: How can luxury hotel brands effectively enhance Generation Z’s booking intentions through integrated marketing communication message strategies? Specifically, using The Luxury Collection as a case study, this research selects three highly representative IMC message appeals in the contemporary luxury travel context—(A) Immersive Cultural Experience, (B) Sustainable Luxury Tourism, and (C) Flexible Stay—and employs an online controlled experiment to systematically compare the effects of these message types on Gen Z consumers’ brand attitude, booking intention, and willingness-to-pay. The findings aim to provide theoretical support and practical insights for luxury hotel brands to develop more resonant and effective IMC strategies amid this generational shift.
1.2. Research Objectives and Significance
This study aims to systematically examine and compare the effects of three distinct IMC messages—A. Immersive Cultural Experience, B. Sustainable Luxury Tourism, and C. Flexible Stay—on the decision-making process of Gen Z consumers in the context of luxury hotels through a controlled experiment. The research will focus on how these message types influence brand attitude, booking intention, and willingness-to-pay, with a specific emphasis on the mediating role of brand attitude.
The significance of this research is twofold, encompassing both theoretical and practical contributions. In terms of theoretical significance, this study enriches the literature on hotel marketing and generational communication by focusing on the under-explored realm of luxury hospitality. It not only addresses the call for more research on Gen Z consumer behavior but also provides new evidence for the applicability of the classic “cognition-affect-behavior” framework (Rosenberg et al., 1960) in contemporary digital marketing contexts by empirically testing the causal mechanisms between message strategies and key psychological (brand attitude) and behavioral variables (booking and payment intentions).
Regarding practical significance, this study is designed to provide actionable insights for “The Luxury Collection” and comparable luxury hotel brands to optimize their marketing content. By identifying the message appeal most effective in driving the booking intentions of Gen Z, the findings will serve as an evidence-based guide for brand managers to make informed decisions when developing IMC strategies targeted at the future dominant consumer cohort. This will ultimately aid in optimizing marketing resource allocation and enhancing communication efficiency and return on investment.
2. Literature Review and Hypothesis Development
2.1. Gen Z Consumer Behavior and Travel Decision-Making
As the first generation to grow up entirely in the digital age, Generation Z exhibits fundamental differences in consumption patterns compared to their predecessors, which has profoundly reshaped the global travel and hospitality landscape (Corbisiero et al., 2022). Understanding Gen Z’s value systems, media habits, and their reinterpretation of the concept of “luxury” is foundational to exploring effective marketing communication strategies for this cohort.
Firstly, Gen Z’s value system is distinctly characterized by an experience-orientation, a quest for authenticity, and a sense of social responsibility. They increasingly allocate their spending towards experiences that promise lasting memories, personal growth, and social sharing, rather than mere material possession (Corbisiero et al., 2022). In a travel context, this translates into a declining interest in superficial “checklist” tourism and a growing preference for deep, authentic connections with destination cultures and communities (He & Timothy, 2024). Concurrently, Gen Z expects brands to take clear stances on ethical and social issues. They often align their consumption choices with their personal values by supporting brands that demonstrate strong environmental practices, fair trade, and community engagement (Djafarova & Foots, 2022; Seyfi et al., 2024). Consequently, for luxury hotel brands, showcasing opulent amenities is no longer sufficient; communicating deeper experiential value and social meaning has become imperative.
Secondly, as digital natives, Gen Z’s media usage and information processing patterns are unique. They are adept at navigating multi-channel, multi-tasking information environments, swiftly filtering and integrating vast amounts of information from social media, online reviews, influencer recommendations, and official brand content. Their travel decision journey is, therefore, highly non-linear, fragmented, and reliant on peer validation (Pricope Vancia et al., 2023). This pattern places heightened demands on IMC: brand messaging must maintain consistency across platforms while also being highly shareable and possessing “social currency” value to stimulate engagement and dissemination within this demographic (Kesgin & Murthy, 2019).
2.2. Integrated Marketing Communications and Message Strategy Theory
Integrated Marketing Communications theory represents a core paradigm in modern marketing. Its essence lies in coordinating all of a brand’s communication channels and messaging to deliver a consistent and reinforced voice, thereby shaping a unified brand perception in a multi-touchpoint environment and effectively enhancing consumers’ brand attitudes and behavioral intentions (Kitchen & Tourky, 2022). In today’s highly fragmented media landscape, the key to successful IMC hinges on ensuring high synergy in themes, value propositions, and expression across different channels, enabling consumers to construct a coherent, rather than conflicting, brand meaning. This is not merely a path to increased communication efficiency but also a foundation for building brand credibility and long-term relationships (Rehman et al., 2022).
Within the IMC framework, message strategy is pivotal to determining communication effectiveness. The fundamental distinction lies between emotional appeals and functional appeals. Emotional appeals aim to evoke audience’s affective responses, fostering emotional associations through resonance, experience, and symbolism, thereby influencing subjective attitudes (Rigby & Lee, 2024). Particularly in digital media environments, emotional appeals can effectively enhance consumer engagement and memory retention, promoting deep brand loyalty (Basha et al., 2025). Conversely, functional appeals focus on conveying the specific, practical advantages of a product or service, such as performance, convenience, and value-for-money, primarily targeting the consumer’s rational evaluation process (Han et al., 2019). These two appeals are not mutually exclusive but are strategically complementary. The optimal strategy depends on brand objectives, consumer characteristics, and the contact situation—for instance, functional information may enhance perceptions of reliability in high-involvement decisions, whereas emotional appeals are often more effective in shaping brand personality and long-term relationships (Aaker, 1991).
2.3. Key Factors Influencing Hotel Booking Intentions
Hotel booking intention, a critical predictor of consumer choice behavior, is shaped by a complex interplay of factors at multiple levels. In the competitive luxury hotel market, brand equity serves as the cornerstone influencing consumer decision-making. Based on Aaker’s seminal framework, brand equity is a multidimensional concept encompassing brand loyalty, brand awareness, perceived quality, brand associations, and other proprietary assets (Aaker, 1991). Within this framework, brand attitude—defined as consumers’ overall and enduring evaluation of a brand’s favorability—constitutes the core affective and cognitive dimension of brand equity (Keller, 1993). Beyond brand-related factors, consumer decision-making is fundamentally a value-based rational process, making perceived value another central lens for understanding booking intentions. Zeithaml defined perceived value as the consumer’s overall assessment of the utility of a product based on perceptions of what is received and what is given (Zeithaml, 1988). It has evolved from a simplistic price-value consideration to a multifaceted construct including functional value (e.g., core service, convenience), emotional value (e.g., pleasure, experience), social value (e.g., status, belonging), and epistemic value (e.g., novelty) (Sánchez-Fernández & Iniesta-Bonillo, 2006). In the experience economy, luxury hotel consumers seek far more than mere accommodation; they pursue emotional satisfaction, self-actualization, and unique memories through their consumption. This comprehensive perceived value is directly translated into market behavior, namely willingness-to-pay. When consumers perceive greater overall benefits from a brand’s value proposition, their willingness-to-pay increases, potentially accepting a brand premium (Dwivedi et al., 2018). In the contemporary travel consumption environment, particularly in the post-pandemic era, heightened consumer risk awareness and desire for control have elevated policy flexibility from a value-added “nice-to-have” to a core value driver (Liu et al., 2025). For Gen Z travelers, who champion autonomy, are risk-averse, and are adept at digital comparison, flexible booking policies have become a fundamental requirement (Morrone et al., 2023). Theoretically, flexible policies primarily operate by reducing consumers’ perceived risk. According to risk aversion theory, as decision uncertainty increases, any measure that reduces uncertainty will boost decision confidence and action intention (Werner, 2016).
2.4. Hypothesis Development
H1: IMC message type will have a significant direct effect on booking intention.
Specifically, it is predicted that the Immersive Cultural Experience message (Group A) will be significantly more effective in enhancing booking intention than both the Sustainable Luxury Tourism message (Group B) and the Flexible Stay message (Group C). This prediction is grounded in the notion that the core value of luxury hotels lies in delivering experiences that transcend functionality and carry high emotional value. The immersive cultural message, by promising unique destination stories and opportunities for self-extension, most directly aligns with Gen Z’s core pursuit of “experiential luxury” (Kitchen & Tourky, 2022), “authenticity” (He & Timothy, 2024), and “social currency” (Kesgin & Murthy, 2019). In contrast, while the sustainability message resonates with their values, it may be less direct than a high-impact experiential message in driving immediate booking behavior. The flexibility message, often perceived as a fundamental risk-reducer or functional prerequisite, offers limited emotional stimulation, and is thus predicted to be the least effective in stimulating the booking intention.
H2: IMC message type will have a significant effect on brand attitude.
Brand attitude, as the core evaluation of a brand, is highly susceptible to the content of external communications. Different IMC message appeals are expected to shape brand attitude through distinct psychological pathways. Drawing on message consistency theory, it is predicted that the three message types, with their unique value propositions, will elicit differentiated cognitive and affective responses from Gen Z consumers, thereby exerting significantly different effects on the formed brand attitude.
H3: Brand attitude will mediate the relationship between IMC message type and booking intention.
This hypothesis constitutes the theoretical core of this study. According to the classic “cognition-affect-behavior” framework, external stimuli often influence behavioral intentions indirectly by first impacting internal psychological variables. It is postulated that the IMC message type will first influence the consumer’s overall evaluation of the brand (brand attitude), and this altered attitude will, in turn, guide subsequent behavioral tendencies (booking intention). In other words, the effect of message type on booking intention is hypothesized to be at least partially transmitted through the activation or alteration of brand attitude.
H4: IMC message type will have a significant effect on willingness-to-pay.
Willingness-to-pay is a direct manifestation of perceived value. By conveying different value propositions, the IMC messages are expected to lead to variations in consumers’ overall perceived value of the hotel service. When consumers perceive higher value from a particular message, their accepted price premium is correspondingly higher. Therefore, it is predicted that the three message types will have a significantly different impact on consumers’ willingness-to-pay.
3. Research Methodology and Design
3.1. Overall Research Design
To test the proposed hypotheses and establish the causal effect of different integrated marketing communication messages on the outcome variables, this study employed an experimental method as the core research strategy. The primary rationale for choosing an experiment lies in its strong advantage in causal inference. By actively manipulating the independent variable and randomly assigning participants to different experimental groups, the experimental method effectively controls extraneous variables, thereby allowing observed differences in the outcome variables to be attributed to the manipulation of the independent variable—a level of rigor difficult to achieve with correlational or survey-based research (Campbell & Stanley, 2015).
Specifically, this study adopted a single-factor, between-subjects design.
1) Group A: Immersive Cultural Experience message
2) Group B: Sustainable Luxury Tourism message
3) Group C: Flexible Stay message
The dependent variables include: brand attitude, booking intention, and willingness-to-pay.
The between-subjects design means that each respondent is randomly assigned to only one of the three experimental groups and exposed to a single experimental stimulus. This design effectively avoids potential carryover effects, fatigue effects, or hypothesis guessing associated with within-subject designs, thereby better safeguarding the internal validity of the experiment.
3.2. Experimental Stimuli Design
Group A: Immersive Cultural Experience
This advertisement was designed to communicate a value proposition centered on deep cultural integration with the destination. The advertisement emphasized immersive cultural experience and destination storytelling.
Group B: Sustainable Luxury Tourism
This advertisement focused on the brand’s commitment to environmental protection and social responsibility, framing the act of staying at the hotel as a contribution to preserving the destination’s environment and culture.
Group C: Flexible Stay
This advertisement highlighted flexibility and a sense of control during travel. The design process for all stimuli was as follows: First, official marketing materials from The Luxury Collection were systematically collected and analyzed. Second, a unified advertisement visual template was created based on this analysis. Finally, differing core copy and taglines were embedded into this consistent template for the three experimental conditions to operationalize the independent variable (message type).
3.3. Variable Measurement and Scales
3.3.1. Dependent Variables
This study included three key dependent variables:
Booking Intention: Measured using a 4-item, 7-point Likert scale adapted from the work of Zeithaml (Liu et al., 2025), anchored from 1 (Strongly Disagree) to 7 (Strongly Agree). A sample item is: “After viewing this advertisement, my likelihood of considering ‘The Luxury Collection’ for a future stay has increased.” A composite score for booking intention was calculated by averaging the four items, with higher scores indicating stronger booking intention.
Brand Attitude: Measured using a 4-item, 7-point semantic differential scale. Respondents evaluated their overall brand perception on four pairs of bipolar adjectives (e.g., “Unattractive-Attractive”). A composite score was calculated by averaging all items, with higher scores reflecting a more favorable brand attitude.
Willingness-to-Pay: Measured using a payment card approach. Respondents were asked: “Assuming you plan a two-night stay at a destination, what is the maximum average price per night you are willing to pay for ‘The Luxury Collection’?” They selected from six ordered options ranging from “Below RMB 1500” to “Above RMB 3501”. Thus, WTP is inherently an ordinal categorical variable.
3.3.2. Control Variables
To account for other potential influences, the following variables were measured:
Demographic Variables: Included age (continuous), gender (categorical), education level (ordinal), and monthly disposable income (ordinal).
Travel Frequency: Measured by a single item: “Approximately how many times have you stayed in luxury hotels (average price ≥ RMB 2000/night) in the past 3 years?” with ordered response categories (e.g., “0 times”, “1 - 2 times”).
Environmental Concern: Measured using a short 4-item, 7-point Likert scale (e.g., “I am an environmentally conscious consumer”). A mean score was calculated, with higher scores indicating greater environmental concern.
The internal consistency reliability (Cronbach’s α) for all multi-item scales was assessed prior to the main analysis.
3.4. Questionnaire Design and Pilot Test
Introduction and Informed Consent: Respondents were first informed about the research purpose, anonymity, and confidentiality, and their consent was obtained.
Screening Questions: Questions regarding year of birth, past and future travel plans, and brand awareness were used to screen for the target population (Chinese Gen Z, potential consumers with travel intent). Respondents failing to meet the criteria were terminated from the survey.
Random Exposure to Experimental Stimulus: The platform’s random assignment function evenly and randomly allocated eligible respondents to one of the three groups (A, B, or C), who were then shown the corresponding advertisement.
Main Variable Measurement: Immediately after viewing the stimulus, the dependent variables (booking intention, willingness-to-pay) and the mediating variable (brand attitude) were measured.
Control Variables and Attention Check: Control variables, including environmental consciousness and luxury travel frequency, were measured subsequently. An attention check question was included to identify inattentive respondents.
Demographic Information: Finally, demographic data such as age, gender, and income were collected.
Prior to the full-scale launch, a pilot test was conducted, yielding 30 valid responses. The objectives of the pilot test were:
Scale Reliability Check: Reliability analysis for the booking intention and brand attitude scales showed Cronbach’s α coefficients all exceeded the acceptable threshold of 0.7, indicating good internal consistency.
Stimulus Check: Open-ended feedback confirmed that the core messages of the three stimuli were correctly understood and perceived as distinct and authentic, validating the effective manipulation of the independent variable.
3.5. Data Collection Procedure
During the formal data collection phase, the questionnaire was distributed via the Wenjuanxing platform. The sampling strategy targeted Chinese Gen Z consumers aged 18 to 25, who had engaged in leisure travel in the past 12 months and planned to travel in the coming 12 months.
A combination of quota sampling and snowball sampling was employed, disseminating the questionnaire through social media channels and university networks. The platform’s screening logic ensured only respondents meeting all demographic and behavioral criteria proceeded to the experimental stage.
The target was set at 600 valid questionnaires. During data cleaning, responses failing the attention check, exhibiting short completion times, or showing patterned responses were excluded. Ultimately, 600 eligible respondents were successfully included in the analysis, randomly and evenly assigned to the three experimental groups (n = 200 per group), ensuring inter-group comparability.
3.6. Data Analysis Methods
Following data collection, analyses were conducted using the SPSS 26.0 statistical software package, proceeding through the following steps:
Step 1: Descriptive Statistics and Reliability/Validity Tests
Initially, descriptive statistics were computed for the valid sample. This included reporting frequency distributions for demographic variables and group assignments, and calculating means and standard deviations for continuous variables to understand sample characteristics and the distribution of key variables.
Subsequently, reliability and validity of the multi-item scales were assessed. Reliability analysis was performed to calculate Cronbach’s alpha coefficients for the booking intention and brand attitude scales, with α > 0.70 considered acceptable for internal consistency. Additionally, Exploratory Factor Analysis was employed to examine construct validity, using Principal Component Analysis with Varimax rotation to confirm the anticipated factor structure of the scales. The Kaiser-Meyer-Olkin measure and Bartlett’s test of sphericity were examined to assess the suitability of the data for factor analysis.
Step 2: Hypothesis Testing
Specific statistical methods were applied to test the research hypotheses:
Testing H1, H2, H4: One-way Analysis of Variance was used to examine the main effect of message type on booking intention (H1), brand attitude (H2), and willingness-to-pay (H4). If the ANOVA indicated significant between-group differences, Tukey’s HSD post hoc tests were conducted for pairwise comparisons to identify which specific groups differed.
Testing H3 (Mediation Effect): The mediation effect of brand attitude in the relationship between message type and booking intention (H3) was tested using the PROCESS macro for SPSS (Model 4). Bootstrapping with 5000 resamples was used to generate an estimate of the indirect effect and its 95% bias-corrected confidence interval. A significant mediation effect is concluded if the confidence interval does not include zero.
Step 3: Supplementary Analysis (Robustness Check)
To check the robustness of the main and mediation effects, and to explore the influence of control variables, a multiple linear regression analysis was performed. Booking intention served as the dependent variable, with predictors including message type, brand attitude, and all control variables. This analysis aimed to test whether the effects of the core independent variables remained significant after accounting for the control variables.
4. Data Analysis and Results
4.1. Sample Structure and Descriptive Statistics
To test the internal consistency of the scale, this study first collected 30 samples for a pre-test of questionnaire reliability and validity, and conducted reliability analysis on the three scales: Booking Intention (BI), Brand Attitude (BA), and Eco-Consciousness (ECO) (see Table 1). The results showed that the Cronbach’s α coefficients of the three scales were all above 0.87, namely 0.91, 0.878, and 0.88 respectively. The standardized Cronbach’s α coefficients were also consistent, indicating that the items of each scale have good internal consistency in measuring the same underlying construct.
Table 1. Reliability analysis.
Scale |
Cronbach’s α |
Standardized Cronbach’s α |
Number of Items |
Sample Size (n) |
Booking Intention (BI) |
0.91 |
0.91 |
4 |
30 |
Brand Attitude (BA) |
0.878 |
0.88 |
4 |
30 |
Eco-Consciousness (ECO) |
0.88 |
0.883 |
4 |
30 |
To further verify the structural rationality of the scales, this study conducted KMO tests and Bartlett’s sphericity tests on three scales (see Table 2). The results showed that the KMO values were all above 0.7 (BI = 0.833, BA = 0.759, ECO = 0.801), indicating that the samples were suitable for factor analysis; at the same time, the approximate chi-square values of the Bartlett’s sphericity test reached significant levels (p < 0.001), suggesting that there were significant correlations among the items of the scales. Based on the results of the comprehensive reliability and validity analysis, these three scales were structurally reasonable and measurement was robust in the pre-test, providing a reliable basis for the subsequent formal questionnaire survey.
Table 2. Validity analysis: KMO & Bartlett test.
Scale |
KMO |
Bartlett’s Test χ2 |
df |
p |
Booking Intention (BI) |
0.833 |
84.84 |
6 |
<0.001*** |
Brand Attitude (BA) |
0.759 |
62.693 |
6 |
<0.001*** |
Eco-Consciousness (ECO) |
0.801 |
62.263 |
6 |
<0.001*** |
***p < 0.001 indicates highly significant results.
After conducting the pre-test for reliability and validity, the questionnaire design in this study was found to be reasonable and effective. Therefore, the questionnaire was publicly released and samples were collected. A total of 658 samples were collected in this study. Samples that answered incorrectly in the “Attention_Check” section were excluded, resulting in the deletion of 35 invalid questionnaires and the retention of 623 valid samples. To control the impact of differences in sample sizes among different experimental groups on the statistical results, the study conducted further balancing processing on the valid samples. Samples were randomly selected from the three experimental groups of “Flexible Stay”, “Immersive Culture Experience”, and “Sustainable Luxury Tourism”, totaling 600 samples, forming a balanced data set for subsequent group comparisons and hypothesis testing.
Based on the questionnaire data after data cleaning, this study generated three core composite variables for the three scales: Booking_Intention was formed by averaging the three items BI1, BI2, BI3 and BI4; Brand_Attitude was formed by averaging the four items Att1 to Att4; and Eco_Consciousness was formed by averaging the two items ECO1 to ECO4.
To test the representativeness and structural characteristics of the samples, this study first conducted descriptive statistical analysis of the demographic variables and willingness to pay (WTP) of the 600 valid samples included in the analysis. The results are shown in Table 3.
Table 3. Sample demographic profile and group distribution.
Name |
Options |
Frequency |
Percentage (%) |
Group |
Flexible Stay |
200 |
33.33 |
Immersive Culture Experience |
200 |
33.33 |
Sustainable Luxury Tourism |
200 |
33.33 |
WTP |
Above 3501 yuan |
125 |
20.83 |
1501 - 2000 yuan |
105 |
17.50 |
2501 - 3000 yuan |
97 |
16.17 |
3001 to 3500 yuan |
94 |
15.67 |
2001 - 2500 yuan |
90 |
15.00 |
Less than 1500 yuan |
89 |
14.83 |
Gender |
Male |
326 |
54.33 |
Female |
274 |
45.67 |
Edu |
Junior college |
140 |
23.33 |
High school and below |
162 |
27.00 |
Undergraduate |
151 |
25.17 |
Master’s degree or above |
147 |
24.50 |
Income |
More than 10,000 yuan |
156 |
26.00 |
6001 to 10,000 yuan |
152 |
25.33 |
Less than 3000 yuan |
149 |
24.83 |
3001 - 6000 yuan |
143 |
23.83 |
Total |
600 |
100 |
In the experimental group, due to the adoption of a balanced sampling design, the sample sizes of the three groups—Flexible Stay, Immersive Culture Experience, and Sustainable Luxury Tourism—were exactly the same, each consisting of 200 people, accounting for 33.33% of the total sample. This laid the foundation for fair inter-group comparisons in the subsequent analysis. In terms of demographic characteristics, the gender distribution was relatively balanced, with 54.33% being male and 45.67% being female. Regarding educational attainment, the distribution across different levels was relatively even, with “high school and below” accounting for the highest proportion (27.00%), followed by “bachelor’s degree” (25.17%) and “master’s degree and above” (24.50%), indicating that the sample had a certain degree of educational diversity. Personal monthly income also showed a relatively uniform distribution, with each income range accounting for approximately 24% - 26%, suggesting that the sample covered different income levels. In terms of core consumption indicator payment willingness (WTP), the distribution of choices across different price ranges was also relatively dispersed. The option “above 3501 yuan” had the highest proportion (20.83%), while the option “below 1500 yuan” had the lowest proportion (14.83%).
After completing the analysis of the sample structure, this study conducted descriptive statistical analysis on the measurement items and composite variables of each core variable to preliminarily understand the central tendency and dispersion of the data, as shown in Table 4.
Table 4. Descriptive statistics of the core variables (N = 600).
Variable name |
N |
Maximum |
Minimum |
Mean |
S.D. |
Booking_Intention |
600 |
7 |
1.25 |
4.621 |
1.252 |
BI1 |
600 |
7 |
1 |
4.607 |
1.476 |
BI2 |
600 |
7 |
1 |
4.555 |
1.52 |
BI3 |
600 |
7 |
1 |
4.532 |
1.448 |
BI4 |
600 |
7 |
1 |
4.79 |
1.335 |
Brand_Attitude |
600 |
7 |
1.25 |
4.522 |
1.287 |
BA1 |
600 |
7 |
1 |
4.472 |
1.491 |
BA2 |
600 |
7 |
1 |
4.525 |
1.446 |
BA3 |
600 |
7 |
1 |
4.53 |
1.458 |
BA4 |
600 |
7 |
1 |
4.563 |
1.486 |
Eco_Consciousness |
600 |
7 |
1.25 |
4.487 |
1.26 |
ECO1 |
600 |
7 |
1 |
4.578 |
1.434 |
ECO2 |
600 |
7 |
1 |
4.477 |
1.459 |
ECO3 |
600 |
7 |
1 |
4.437 |
1.567 |
ECO4 |
600 |
7 |
1 |
4.455 |
1.558 |
Age |
600 |
25 |
18 |
21.658 |
2.286 |
From the descriptive statistical analysis results table, it can be seen that the means of all the comprehensive variables are at a moderately high level. The mean of Booking_Intention (M = 4.621, SD = 1.252) is the highest, indicating that the sample as a whole has a relatively positive consumption intention for tourism products. The means of Brand_Attitude (M = 4.522, SD = 1.287) and Eco_Consciousness (M = 4.487, SD = 1.260) are similar, showing that the respondents have relatively positive evaluations of the brand image and environmental issues. The mean of the control variable Age is 21.658 years (SD = 2.286), with a minimum value of 18 years and a maximum value of 25 years. This is in line with the preset of this study that the main survey subjects are young people, and the age structure of the sample is relatively concentrated.
4.2. Reliability and Validity Analysis Results
To ensure the scientificity and reliability of the measurement tools, this study employed Cronbach’s α coefficient as the test indicator for internal consistency reliability, conducting reliability analysis on each scale. The results are presented in Table 5.
Table 5. The reliability analysis results of each scale (N = 600).
|
The average value after deleting items |
The variance after deleting the item |
The correlation between the deleted items and the overall population after deletion |
Cronbach’s α coefficient after deleting the item |
Cronbach’s α |
Number of items |
N |
BI1 |
13.877 |
14.302 |
0.77 |
0.851 |
0.888 |
4 |
600 |
BI2 |
13.928 |
13.846 |
0.788 |
0.844 |
BI3 |
13.952 |
14.63 |
0.754 |
0.857 |
BI4 |
13.693 |
15.739 |
0.713 |
0.872 |
BA1 |
13.618 |
15.255 |
0.775 |
0.868 |
0.898 |
4 |
600 |
BA2 |
13.565 |
15.785 |
0.751 |
0.877 |
BA3 |
13.56 |
15.379 |
0.787 |
0.864 |
BA4 |
13.527 |
15.235 |
0.781 |
0.866 |
ECO1 |
13.368 |
15.592 |
0.683 |
0.825 |
0.857 |
4 |
600 |
ECO2 |
13.47 |
15.134 |
0.715 |
0.812 |
ECO3 |
13.51 |
14.474 |
0.709 |
0.814 |
ECO4 |
13.492 |
14.644 |
0.697 |
0.819 |
The results of the reliability analysis of each scale show that the Cronbach’s α coefficient of the Consumption Intention Scale (BI) is 0.888, that of the Brand Attitude Scale (BA) is 0.898, and that of the Environmental Consciousness Scale (ECO) is 0.857, all of which are significantly higher than the general threshold of 0.70. This indicates that the items within each scale have a high degree of consistency. From the “item-total correlation coefficient” (i.e., the correlation coefficient after deleting the item with respect to the total score of the scale), the correlation coefficients of all items are between 0.683 and 0.788, all of which are greater than the recommended standard of 0.50. From the “Cronbach’s α coefficient after deleting the item” column, it can be found that regardless of deleting any item, the α coefficient of each scale does not show a significant improvement, but is generally lower or equal to the overall α coefficient of the scale. In conclusion, the reliability test measurement results of the three scales designed in this study are reliable and have passed the scale reliability analysis.
To verify the structural rationality of the formal questionnaire, this study conducted factor analysis on the three scales: Booking Intention (BI), Brand Attitude (BA), and Eco-Consciousness (ECO). According to the results in Tables 6-8, the KMO values of the three scales are all higher than 0.79 (BI = 0.841, BA = 0.848, ECO = 0.794), and the approximate chi-square values of the Bartlett’s sphericity test are all significant (p < 0.001), indicating that the sample is suitable for factor analysis, and there is a significant correlation between the scale items, and the structure is reasonable.
Table 6. Factor analysis of Booking Intention (BI) scale.
Component |
Eigenvalue |
% of Variance |
Cumulative % |
Rotated Eigenvalue |
% of Variance (Rotated) |
Cumulative % (Rotated) |
KMO |
Bartlett’s Test of Sphericity |
1 |
2.998 |
74.96% |
74.96% |
2.118 |
52.95% |
52.95% |
0.841 |
Approx. χ2 = 1326.125,
df = 6, p < 0.001*** |
2 |
0.399 |
9.97% |
84.92% |
1.279 |
31.98% |
84.92% |
|
|
3 |
0.316 |
7.91% |
92.83% |
|
|
|
|
|
4 |
0.287 |
7.17% |
100% |
|
|
|
|
|
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 7. Factor analysis of Brand Attitude (BA) scale.
Component |
Eigenvalue |
% of Variance |
Cumulative % |
Rotated Eigenvalue |
% of Variance (Rotated) |
Cumulative % (Rotated) |
KMO |
Bartlett’s Test of Sphericity |
1 |
3.065 |
76.62% |
76.62% |
2.12 |
53.00% |
53.00% |
0.848 |
Approx.χ2 = 1425.017,
df = 6, p < 0.001*** |
2 |
0.343 |
8.59% |
85.21% |
1.288 |
32.21% |
85.21% |
|
|
3 |
0.311 |
7.78% |
93.00% |
|
|
|
|
|
4 |
0.28 |
7.00% |
100% |
|
|
|
|
|
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 8. Factor analysis of Eco-Consciousness (ECO) scale.
Component |
Eigenvalue |
% of Variance |
Cumulative % |
Rotated Eigenvalue |
% of Variance (Rotated) |
Cumulative % (Rotated) |
KMO |
Bartlett’s Test of Sphericity |
1 |
2.802 |
70.04% |
70.04% |
1.688 |
42.21% |
42.21% |
0.794 |
Approx.χ2 = 1080.042,
df = 6,
p < 0.001*** |
2 |
0.541 |
13.53% |
83.57% |
1.654 |
41.36% |
83.57% |
|
|
3 |
0.331 |
8.29% |
91.85% |
|
|
|
|
|
4 |
0.326 |
8.15% |
100% |
|
|
|
|
|
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
In terms of factor extraction, the first factor of each scale accounted for the majority of the variance (BI = 52.95%, BA = 53.00%, ECO = 42.21%). After rotation, the cumulative variance explained by the first two factors exceeded 80%, and the total variance explained rate reached or approached 100%. This indicates that the items of the three scales can well reflect their respective latent constructs, and the structural validity of the scales is high, making them suitable for use as measurement tools in formal surveys.
3. Results of One-Way ANOVA Test
To test whether the different experimental treatments have a significant impact on the core variables, this study used “group” as the independent variable and “consumption intention”, “brand attitude” and “payment willingness” as the dependent variables. A one-way ANOVA was conducted, and the results are shown in Table 9.
Table 9. Result table of one-way analysis of variance.
Variable name |
Variable value |
N |
Mean |
S. D. |
F |
p |
Booking_Intention |
Flexible Stay |
200 |
3.569 |
0.943 |
245.357 |
0.000*** |
Immersive Culture Experience |
200 |
5.625 |
1.106 |
Sustainable Luxury Tourism |
200 |
4.669 |
0.69 |
Brand_Attitude |
Flexible Stay |
200 |
3.572 |
0.941 |
218.183 |
0.000*** |
Immersive Culture Experience |
200 |
5.606 |
1.109 |
Sustainable Luxury Tourism |
200 |
4.389 |
0.875 |
WTP_Group |
Flexible Stay |
200 |
2.705 |
1.644 |
92.202 |
0.000*** |
Immersive Culture Experience |
200 |
4.75 |
1.556 |
Sustainable Luxury Tourism |
200 |
3.43 |
1.369 |
Note: ***, **, and * represent significance levels of 1%, 5%, and 10% respectively.
Table 9 presents the score differences and significance test results of the three core variables among different experimental groups (Flexible Stay, Immersive Culture Experience, Sustainable Luxury Tourism). The results of the variance analysis indicate that the differences between the groups are statistically extremely significant. Specifically, the differences in consumption intention are significant (F = 245.357, p < 0.001), the differences in brand attitude are significant (F = 218.183, p < 0.001), and the differences in payment willingness are also significant (F = 92.202, p < 0.001). All p-values are less than 0.001, indicating that different experimental scenarios have a highly significant impact on consumers’ consumption intention, brand evaluation, and payment willingness.
To test the degree of differences in key behavioral and attitude variables among different groups, this paper conducts an analysis of the differences between groups for Booking Intention, Brand Attitude, and WTP Group. Table 10 summarizes the sum of squared differences between groups, total deviation, partial η2 (Partial η2), and Cohen’s f effect size indicators for each analysis index to comprehensively evaluate the explanatory power and actual impact of different groups on each dependent variable.
Table 10. Analysis of the differential effects of different groups in terms of booking intentions, brand attitudes and payment willingness.
Analysis item |
Inter-group difference |
Total deviation |
Partial η2 |
Cohen’s f |
Booking_Intention |
423.505 |
938.74 |
0.451 |
0.907 |
Brand_Attitude |
418.981 |
992.196 |
0.422 |
0.855 |
WTP_Group |
430.003 |
1822.118 |
0.236 |
0.556 |
From the results in Table 10, it can be seen that there are significant differences in all three variables among different groups. Among them, the partial η2 for the intention to book is 0.451, and Cohen’s f reaches 0.907, indicating that the group factors have a very strong explanatory power and a considerable practical impact effect on the intention to book. The partial η2 for brand attitude is 0.422, and Cohen’s f is 0.855, also showing a significant and strong effect level, indicating that there are obvious differences in the cognition and evaluation of brand attitude among different groups. In contrast, the partial η2 for the payment willingness group is 0.236, and Cohen’s f is 0.556, although the effect intensity is relatively low, it still reaches a moderately large effect level, indicating that the group factors still have a certain influence on consumers’ payment willingness.
Based on confirming the significant overall differences among different groups, this paper further adopts the post hoc multiple comparison method to test the specific differences in the intention to book, brand attitude, and payment willingness groups (WTP Group) among different tourism context types (Flexible Stay, Immersive Culture Experience, and Sustainable Luxury Tourism). Table 11 shows the means, mean differences (I − J), and significance levels (p values) among each group to clearly identify the direction and magnitude of the differences between different contexts (Figure 1).
From the results in Table 11, it can be seen that there is a significant difference between Flexible Stay and Immersive Culture Experience in terms of booking intention, and the former is significantly lower than the latter (Difference = −2.056, p < 0.001); at the same time, the booking intention of Flexible Stay is also significantly lower than that of Sustainable Luxury Tourism (Difference = −1.100, p < 0.001), while Immersive Culture Experience is significantly higher than Sustainable Luxury Tourism (Difference = 0.956, p < 0.001).
Table 11. Post-hoc multiple comparison results of each variable in different tourism scenarios.
|
(I) Name |
(J) Name |
(I) Average value |
(J) Average value |
Difference (I − J) |
p |
Booking_Intention |
Flexible Stay |
Immersive Culture Experience |
3.569 |
5.625 |
−2.056 |
0.000*** |
Flexible Stay |
Sustainable Luxury Tourism |
3.569 |
4.669 |
−1.1 |
0.000*** |
Immersive Culture Experience |
Sustainable Luxury Tourism |
5.625 |
4.669 |
0.956 |
0.000*** |
Brand_Attitude |
Flexible Stay |
Immersive Culture Experience |
3.572 |
5.606 |
−2.034 |
0.000*** |
Flexible Stay |
Sustainable Luxury Tourism |
3.572 |
4.389 |
−0.816 |
0.000*** |
Immersive Culture Experience |
Sustainable Luxury Tourism |
5.606 |
4.389 |
1.218 |
0.000*** |
WTP_Group |
Flexible Stay |
Immersive Culture Experience |
2.705 |
4.75 |
−2.045 |
0.000*** |
Flexible Stay |
Sustainable Luxury Tourism |
2.705 |
3.43 |
−0.725 |
0.000*** |
Immersive Culture Experience |
Sustainable Luxury Tourism |
4.75 |
3.43 |
1.32 |
0.000*** |
Note: ***, **, and * represent significance levels of 1%, 5%, and 10% respectively.
Figure 1. Analysis of variance comparison chart.
In terms of brand attitude, the comparison results show a consistent trend, that is, Immersive Culture Experience has the highest brand attitude evaluation, followed by Sustainable Luxury Tourism, while Flexible Stay has the lowest score. The differences between each group are all at a significant level (p < 0.001).
In terms of payment willingness, Immersive Culture Experience is also significantly higher than the other two scenarios, with the difference between it and Flexible Stay being the most significant (Difference = 2.045, p < 0.001), while the payment willingness level of Sustainable Luxury Tourism is between the two. Overall, the immersive culture experience shows the best effect in all three indicators, indicating its obvious advantage in enhancing consumer attitudes and behavioral intentions.
In conclusion, the hypotheses H1, H2, and H4 in this study are all valid, and there are significant differences in the impact of the three sets of information on booking intention, brand attitude, and payment willingness.
4.3. Mediating Effect Test Results
To test the mediating effect of Brand Attitude between different tourism context types and Booking Intention, this study uses multiple regression methods to test the mediating effect. This research constructs three regression models: the first step tests the total effect of the independent variable on the dependent variable; the second step tests the effect of the independent variable on the mediating variable; the third step, with the introduction of the mediating variable, tests the direct effect of the independent variable on the dependent variable and the effect of the mediating variable. At the same time, to improve the robustness of the mediating effect test results, this study uses the Bootstrap method for 5000 repeated sampling to obtain more stable estimation results.
In the mediating effect analysis of this study, this paper processes the tourism context type variable using Dummy Variables and sets Flexible Stay as the reference group, constructing two dummy variables for “Immersive Culture Experience” and “Sustainable Luxury Tourism”, and the regression coefficients reflect the marginal impact of each tourism context on brand attitude and booking intention relative to Flexible Stay, while controlling other variables unchanged. The reason for choosing Flexible Stay as the reference group is mainly because its functionality and experience intensity are relatively low, facilitating the highlighting of the difference effects of immersive culture experience and sustainable luxury tourism in terms of emotional and value perception. In terms of controlling variable settings, this paper incorporates variables such as consumption frequency (Freq_Lux), gender (Gender), age (Age), and income level (Income Group). Among them, the gender variable is handled as a dummy variable. While the consumption frequency and income level are sequential variables with clear hierarchical order, to maintain the integrity of variable information and avoid excessive parameterization, this paper converts the sequential variables into numerical values based on their original hierarchical order and introduces them as continuous variables into the regression model, thereby reflecting the linear influence of their hierarchical changes on the dependent variable.
From the regression results in Table 12, it can be seen that without introducing the mediating variables, different tourism context types have a significant positive impact on the booking intention. Among them, compared with the baseline group, Immersive Culture Experience and Sustainable Luxury Tourism significantly enhance consumers’ booking intention (p < 0.001). At the same time, the control variables, Eco Consciousness, shows a very strong positive impact on the booking intention (β = 0.820, p < 0.001), and the model’s explanatory power is high (R2 = 0.817), indicating that the independent variables have a significant overall effect on the booking intention, providing a prerequisite for the mediation effect test.
Table 12. The regression test results of the mediating effect of brand attitude between the type of tourism context and booking intention-1.
|
Booking_Intention |
Coefficient |
Standard error |
t |
p |
Standardization coefficient |
c |
0.927 |
0.253 |
3.671 |
0.000*** |
- |
Group_Sustainable Luxury Tourism |
0.373 |
0.06 |
6.226 |
0.000*** |
0.141 |
Group_Immersive Culture Experience |
0.35 |
0.075 |
4.652 |
0.000*** |
0.132 |
Eco_Consciousness |
0.815 |
0.024 |
33.443 |
0.000*** |
0.82 |
Freq_Lux |
−0.019 |
0.027 |
−0.708 |
0.479 |
−0.013 |
Gender_Male |
0.061 |
0.044 |
1.383 |
0.167 |
0.024 |
Age |
−0.002 |
0.01 |
−0.161 |
0.873 |
−0.003 |
Income_Group |
−0.05 |
0.027 |
−1.848 |
0.065* |
−0.034 |
R2 |
0.817 |
Ad-R2 |
0.814 |
F |
F(7, 592) = 376.645, p = 0.000*** |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 13 further examines the impact of tourism context types on brand attitude. The results show that the immersive cultural experience has a significant positive impact on brand attitude (β = 0.145, p < 0.001), while sustainable luxury tourism has no significant positive impact on brand attitude (p > 0.05). In the control variables, ecological consciousness also has a significant positive effect on brand attitude (β = 0.771, p < 0.001), indicating that different tourism contexts can influence consumers’ brand evaluation and emotional attitude, laying the foundation for subsequent behavioral intention changes, and the model is overall significant (F = 251.814, p < 0.001), with strong explanatory power (R2 = 0.749).
Table 13. The regression test results of the mediating effect of brand attitude between tourism context type and booking intention-2.
|
Brand_Attitude |
Coefficient |
Standard error |
t |
p |
Standardization coefficient |
c |
1.003 |
0.304 |
3.296 |
0.001*** |
- |
Group_Sustainable Luxury Tourism |
0.128 |
0.072 |
1.771 |
0.077* |
0.047 |
Group_Immersive Culture Experience |
0.396 |
0.091 |
4.369 |
0.000*** |
0.145 |
Eco_Consciousness |
0.787 |
0.029 |
26.849 |
0.000*** |
0.771 |
Freq_Lux |
−0.036 |
0.033 |
−1.097 |
0.273 |
−0.024 |
Gender_Male |
−0.049 |
0.053 |
−0.911 |
0.363 |
−0.019 |
Age |
0.007 |
0.012 |
0.606 |
0.545 |
0.013 |
Income_Group |
−0.07 |
0.032 |
−2.168 |
0.031** |
−0.046 |
R2 |
0.749 |
Ad-R2 |
0.745 |
F |
F(7, 592) = 251.814, p = 0.000*** |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
In Table 14, after introducing brand attitude as a mediating variable into the booking intention regression model, brand attitude shows a significant positive impact on booking intention (β = 0.393, p < 0.001), indicating that brand attitude is an important psychological mechanism affecting consumers’ booking decisions. The model’s explanatory power further improves (R2 = 0.855), showing that the fitting effect of the model has significantly improved after introducing the mediating variable.
Table 14. The regression test results of the mediating effect of brand attitude between tourism context type and booking intention-3.
|
Booking_Intention |
Coefficient |
Standard error |
t |
p |
Standardization coefficient |
c |
0.544 |
0.227 |
2.402 |
0.017** |
- |
Group_Sustainable Luxury Tourism |
0.325 |
0.053 |
6.073 |
0.000*** |
0.122 |
Group_Immersive Culture Experience |
0.199 |
0.068 |
2.926 |
0.004*** |
0.075 |
Eco_Consciousness |
0.514 |
0.032 |
15.938 |
0.000*** |
0.517 |
Freq_Lux |
−0.006 |
0.024 |
−0.229 |
0.819 |
−0.004 |
Gender_Male |
0.08 |
0.039 |
2.026 |
0.043** |
0.032 |
Age |
−0.004 |
0.009 |
−0.494 |
0.621 |
−0.008 |
Income_Group |
−0.023 |
0.024 |
−0.954 |
0.341 |
−0.016 |
Brand_Attitude |
0.382 |
0.03 |
12.595 |
0.000*** |
0.393 |
R2 |
0.855 |
Ad-R2 |
0.853 |
F |
F(8, 591) = 437.143, p = 0.000*** |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Based on the stepwise regression analysis, this paper further uses the Bootstrap method to directly test the mediation effect to improve the robustness of the estimation results. Table 15 reports the results of the mediation path test of different tourism context types influencing booking intention through brand attitude, including the total effect (c), path a, the indirect effect of the independent variable through the mediating variable on the dependent variable (a × b), Bootstrap standard error, z value, significance level, and 95% Bootstrap confidence interval. At the same time, the direct effect after controlling the mediating variable (c’) is reported to determine the existence and type of the mediation effect. The Bootstrap sampling times are set at 5000 times (Figure 2).
Table 15. The mediating effect of brand attitude between tourism context type and booking intention: Bootstrap test results.
Path |
c |
a |
b |
a × b |
a × b (Boot SE) |
a × b (z) |
a × b(95% Boot CI) |
c’ |
Inspection conclusion |
Group_Sustainable Luxury Tourism → Brand_Attitude → Booking_Intention |
0.373 (0.000***) |
0.128 (0.077*) |
0.382 (0.000***) |
0.049 (0.091*) |
0.029 |
1.695 |
[−0.008 - 0.104] |
0.325 (0.000***) |
The mediating effect is not significant |
Group_Immersive Culture Experience → Brand_Attitude → Booking_Intention |
0.350 (0.000***) |
0.396 (0.000***) |
0.382 (0.000***) |
0.151 (0.000***) |
0.04 |
3.774 |
[0.076, 0.230] |
0.199 (0.004***) |
Partialmediatingrole |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Figure 2. The mediating effect of brand attitude between tourism context type and booking intention: Bootstrap test results.
From the results in Table 15, it can be seen that different tourism context types show significant differences in influencing booking intention through brand attitude. For the sustainable luxury tourism context, the total effect (c = 0.373, p < 0.001) and the direct effect (c’ = 0.325, p < 0.001) are both significant, but the effect of path a is not significant (p > 0.05), and the Bootstrap confidence interval of the indirect effect a × b includes 0, indicating that brand attitude does not form a significant mediating effect in this path. Therefore, brand attitude does not play a significant mediating role between the sustainable luxury tourism context and booking intention.
In contrast, in the immersive cultural experience context, both path a and path b reach significant levels (p < 0.001), the indirect effect a × b is 0.151, the Bootstrap z value is 3.774, and the 95% Bootstrap confidence interval does not include 0 (0.076 - 0.230), indicating that the mediating effect is significant. At the same time, after controlling for brand attitude, the direct effect of the independent variable on booking intention is still significant (c’ = 0.199, p < 0.01), but it is significantly weaker compared to the total effect, indicating that brand attitude plays a partial mediating role between the immersive cultural experience and booking intention.
Overall, brand attitude has context dependence in influencing booking intention through different tourism context types, and only in the immersive cultural experience context does a stable and significant mediating mechanism form. This result indicates that immersive experience indirectly promotes consumer behavioral intention through shaping consumers’ brand cognition and emotional attitude.
To further test the mediating effect of Brand Attitude between different tourism context types and consumers’ willingness to pay (WTP Group), this paper uses the stepwise regression method combined with the Bootstrap method for mediating effect analysis. Specifically, Table 16 examines the total effect of tourism context types on WTP (c path), Table 17 examines the impact of tourism context types on brand attitude (a path), Table 18 examines the direct effect of tourism context types on WTP (c’ path) and the influence of the mediating variable on the dependent variable (b path), and Table 19 summarizes the indirect effect test results based on Bootstrap (5000 sampling), to determine the significance and type of the mediating effect (Figure 3).
Table 16. The test results of the mediating effect of brand attitude between tourism context type and payment willingness-1.
|
WTP_Group |
Coefficient |
Standard error |
t |
p |
Standardization coefficient |
c |
0.333 |
0.591 |
0.563 |
0.574 |
− |
Group_Sustainable Luxury Tourism |
−0.096 |
0.14 |
−0.684 |
0.494 |
−0.026 |
Group_Immersive Culture Experience |
0.109 |
0.176 |
0.619 |
0.536 |
0.029 |
Eco_Consciousness |
0.927 |
0.057 |
16.261 |
0.000*** |
0.669 |
Freq_Lux |
−0.084 |
0.064 |
−1.328 |
0.185 |
−0.041 |
Gender_Male |
0.015 |
0.104 |
0.148 |
0.882 |
0.004 |
Age |
−0.025 |
0.023 |
−1.095 |
0.274 |
−0.033 |
Income_Group |
−0.024 |
0.063 |
−0.378 |
0.705 |
−0.012 |
R2 |
0.483 |
Ad-R2 |
0.477 |
F |
F(7, 592) = 78.961, p = 0.000*** |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 17. The test results of the mediating effect of brand attitude between tourism context type and payment willingness-2.
|
Brand_Attitude |
Coefficient |
Standard error |
t |
p |
Standardization coefficient |
c |
1.003 |
0.304 |
3.296 |
0.001*** |
- |
Group_Sustainable Luxury Tourism |
0.128 |
0.072 |
1.771 |
0.077* |
0.047 |
Group_Immersive Culture Experience |
0.396 |
0.091 |
4.369 |
0.000*** |
0.145 |
Eco_Consciousness |
0.787 |
0.029 |
26.849 |
0.000*** |
0.771 |
Freq_Lux |
−0.036 |
0.033 |
−1.097 |
0.273 |
−0.024 |
Gender_Male |
−0.049 |
0.053 |
−0.911 |
0.363 |
−0.019 |
Age |
0.007 |
0.012 |
0.606 |
0.545 |
0.013 |
Income_Group |
−0.07 |
0.032 |
−2.168 |
0.031** |
−0.046 |
R2 |
0.749 |
Ad-R2 |
0.745 |
F |
F(7, 592) = 251.814, p = 0.000*** |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 18. The test results of the mediating effect of brand attitude between tourism context type and payment willingness-3.
|
WTP_Group |
Coefficient |
Standard error |
t |
p |
Standardization coefficient |
c |
0.016 |
0.589 |
0.027 |
0.979 |
- |
Group_Sustainable Luxury Tourism |
−0.136 |
0.139 |
−0.982 |
0.326 |
−0.037 |
Group_Immersive Culture Experience |
−0.016 |
0.177 |
−0.091 |
0.928 |
−0.004 |
Eco_Consciousness |
0.678 |
0.084 |
8.089 |
0.000*** |
0.49 |
Freq_Lux |
−0.073 |
0.063 |
−1.163 |
0.245 |
−0.035 |
Gender_Male |
0.031 |
0.103 |
0.3 |
0.765 |
0.009 |
Age |
−0.027 |
0.022 |
−1.209 |
0.227 |
−0.035 |
Income_Group |
−0.002 |
0.062 |
−0.026 |
0.979 |
−0.001 |
Brand_Attitude |
0.316 |
0.079 |
4.006 |
0.000*** |
0.233 |
R2 |
0.497 |
Ad-R2 |
0.489 |
F |
F(8, 591) = 72.853, p = 0.000*** |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Table 19. The mediating effect of brand attitude between tourism context type and payment willingness: Bootstrap test results.
Path |
c |
a |
b |
a × b |
a × b (Boot SE) |
a × b (z) |
a × b (95% Boot CI) |
c’ |
Inspection conclusion |
Group_Sustainable Luxury Tourism → Brand_Attitude → WTP_Group |
−0.096 (n.s.) |
0.128 (0.077*) |
0.316 (0.000***) |
0.040 (0.117) |
0.026 |
1.569 |
[−0.003 - 0.101] |
−0.136 (0.326) |
Themediating effect is not significant |
Group_Immersive Culture Experience → Brand_Attitude → WTP_Group |
0.109 (n.s.) |
0.396 (0.000***) |
0.316 (0.000***) |
0.125 (0.004***) |
0.044 |
2.866 |
[0.058, 0.234] |
−0.016 (0.928) |
Complete intermediary |
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
Figure 3. The mediating effect of brand attitude between tourism context type and payment willingness: Bootstrap test results.
From the regression results in Table 16, it can be seen that the direct impact of different tourism context types on consumers’ willingness to pay is not significant. Whether it is the sustainable luxury tourism context or the immersive cultural experience context, their regression coefficients did not reach the significance level (p > 0.05), indicating that without considering the mediating variables, the tourism context type itself is difficult to directly explain the differences in consumers’ willingness to pay. Ecological awareness has a significant and strong positive impact on willingness to pay (β = 0.669, p < 0.001), indicating that consumers’ value factors play a key role in the payment decision.
The results in Table 17 show that tourism context types have a significant impact on brand attitude, among which the positive effect of immersive cultural experience on brand attitude is particularly significant (β = 0.145, p < 0.001), while sustainable luxury tourism has no significant effect on brand attitude.
In Table 18, after introducing brand attitude into the payment willingness model, brand attitude has a significant positive impact on payment willingness (β = 0.233, p < 0.001), while the direct effects of the two types of tourism context types are still not significant, and the regression coefficients have further weakened. This result indicates that the impact of tourism context types on payment willingness is not achieved through a direct path, but mainly acts indirectly through the psychological mechanism of brand attitude. At the same time, the explanatory power of the model has increased from R2 = 0.483 in Table 11 to 0.497, indicating that the introduction of mediating variables improves the model fitting effect.
Further combining the Bootstrap mediating effect test results in Table 19, it can be shown that in the sustainable luxury tourism context, the indirect effect a × b did not pass the significance test, and its 95% Bootstrap confidence interval included 0, indicating that brand attitude did not play a significant mediating role in this path. In the immersive cultural experience context, the indirect effect a × b was significantly established (a × b = 0.125, p < 0.01), and the 95% Bootstrap confidence interval did not include 0. At the same time, after controlling for brand attitude, the direct effect (c’) was not significant, indicating that brand attitude played a complete mediating role between immersive cultural experience and payment willingness. The research results show that the impact of tourism context types on consumers’ willingness to pay is mainly achieved through the mediating mechanism of brand attitude, and this mediating effect is only significant in the immersive cultural experience context.
In summary, the hypothesis H3 in the research assumption is established. Brand attitude plays a mediating role between message type and booking intention, and payment willingness. In the mediating role of brand attitude between message type and booking intention, with Flexible Stay as the control group, brand attitude does not have a significant mediating effect between Sustainable Luxury Tourism and booking intention, and brand attitude has a significant partial mediating effect between Immersive Cultural Experience and booking intention. In the mediating role of brand attitude between message type and payment willingness, with Flexible Stay as the control group, brand attitude does not have a significant mediating effect between Sustainable Luxury Tourism and payment willingness, and brand attitude has a significant complete mediating effect between Immersive Cultural Experience and payment willingness.
4.4. Robustness Test Results
To test the robustness of the main regression results, this paper uses Booking_Intention as the dependent variable and conducts a multiple linear regression analysis. The model includes tourism context type, Brand_Attitude, Eco_Consciousness, Luxurious Travel Frequency (Freq_Lux), age, gender, and income level as independent variables. The results show that the model is overall significant, with F = 295.944, p < 0.001, R2 = 0.860, Adjusted R2 = 0.860, and the variance inflation factor (VIF) is all below the commonly accepted threshold, indicating no multicollinearity problem. This suggests that the model can well explain the differences in booking intentions and has high robustness (Table 20).
Table 20. Robustness check results (Booking intention as the dependent variable).
|
B |
SE |
β |
t |
p |
VIF |
R2 |
Ad-R2 |
F |
c |
0.51 |
0.28 |
- |
1.83 |
0.067* |
- |
0.86 |
0.86 |
F = 295.944,
p = 0.000*** |
Group_Sustainable Luxury Tourism |
0.37 |
0.06 |
0.14 |
6.73 |
0.000*** |
1.80 |
Group_Immersive Culture Experience |
0.27 |
0.07 |
0.10 |
3.82 |
0.000*** |
2.98 |
Brand_Attitude |
0.38 |
0.03 |
0.39 |
12.47 |
0.000*** |
4.00 |
Eco_Consciousness |
0.51 |
0.03 |
0.51 |
15.77 |
0.000*** |
4.33 |
Age |
0.00 |
0.01 |
−0.01 |
−0.41 |
0.682 |
1.01 |
Freq_Lux_1 - 2 times |
−0.11 |
0.19 |
−0.03 |
−0.56 |
0.575 |
14.90 |
Freq_Lux_3 to 5 times |
−0.12 |
0.19 |
−0.04 |
−0.61 |
0.542 |
22.01 |
Freq_Lux_6 times or more |
−0.06 |
0.20 |
−0.02 |
−0.31 |
0.758 |
24.92 |
Gender_Male |
0.08 |
0.04 |
0.03 |
1.95 |
0.051* |
1.02 |
Income_6001- 10,000 yuan |
0.05 |
0.05 |
0.02 |
1.13 |
0.258 |
1.38 |
Income is greater than 10,000 yuan |
0.01 |
0.05 |
0.00 |
0.16 |
0.874 |
1.36 |
Income: Less than 3,000 yuan |
0.28 |
0.16 |
0.05 |
1.75 |
0.080* |
2.97 |
B = unstandardized coefficient; SE = standard error; β = standardized coefficient. Reference groups: Flexible Stay (tourism type), female (gender), lowest frequency group, and 3001 - 6000 yuan (income). Model fit statistics: R2 = 0.858, Adjusted R2 = 0.855, F = 295.944, p < 0.001. Note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively.
From the result table, it can be seen that consistent with the main regression results, sustainable luxury tourism (β = 0.140, p < 0.001) and immersive cultural experiences (β = 0.102, p < 0.001) have both had a significant positive impact on booking intentions compared to the reference group. Brand attitude has a significant positive effect on booking intentions (β = 0.390, p < 0.001).
Regarding the control variables, environmental awareness has a significant positive effect on booking intentions (β = 0.510, p < 0.001), indicating that environmental value is a key factor in booking intentions. Age and frequency of luxury travel did not show significant effects statistically, and their coefficients were stable in size and direction. Variables related to income were mostly not significant, except for the lowest income group, whose effect was weakly significant (p < 0.10).
5. Discussion and Conclusion
5.1. Summary of Research Findings
First, regarding direct effects, the message type had a highly significant main effect on booking intention, brand attitude, and willingness-to-pay. Post-hoc comparisons clearly revealed that the Immersive Cultural Experience group (Group A) scored significantly higher on all three key dependent variables than both the Sustainable Luxury Tourism group (Group B) and the Flexible Stay group (Group C). This indicates that, compared to emphasizing functional benefits or social values, deep cultural experiences that provide emotional resonance and self-extension value are the most effective messaging strategy for engaging Gen Z’s decision-making.
Second, concerning the underlying mechanism, brand attitude played a critical mediating role. However, this mediating effect was context-dependent. For the Immersive Cultural Experience message, brand attitude acted as a partial mediator in its influence on booking intention and a full mediator in its influence on willingness-to-pay. This suggests that while the cultural experience partly drives booking behavior by enhancing brand favorability, its entire effect on stimulating willingness-to-pay is completely dependent on shaping a positive brand attitude. In contrast, brand attitude did not serve as a significant mediator in the path through which the Sustainable Luxury message influenced consumer decisions.
In conclusion, the core findings of this study validate the superior efficacy of the “experience economy” and “narrative communication” in engaging Gen Z consumers. For luxury hotel brands aiming to build deep relationships with this future core demographic, shifting the focus of marketing communication from traditional functional appeals or generalized responsibility claims to constructing unique, authentic, and immersive cultural stories is a key pathway to achieving differentiation and driving business growth.
5.2. Theoretical Implications Discussion
First, regarding why the immersive cultural experience message emerged as the most effective, the results provide a compelling explanation. Its superiority is highly consistent with Gen Z’s core psychological motivations for self-extension and social currency. Unlike messages emphasizing functional convenience or abstract responsibility, the immersive cultural message, by narrating unique destination stories and offering participative experiences, constructs a “shareable experience” for Gen Z consumers that enriches their personal identity narrative and expands their life horizons. This experience fulfills their intrinsic desire for “authenticity” and serves as valuable “social currency”, easily displayed and shared on social media, thereby reinforcing their social identity and personal value. Consequently, this message type transcends mere functional or value propositions, directly addressing Gen Z’s deeper emotional and social needs, resulting in a more potent effect on driving brand attitude and behavioral intentions.
Second, concerning the mediating role of brand attitude, the results strongly support the applicability and boundary conditions of the classic “cognition-affect-behavior” model in the context of contemporary digital marketing for luxury hotels. The finding that brand attitude played a partial and full mediating role in the paths from the immersive cultural message to booking intention and willingness-to-pay, respectively, indicates that for messages capable of eliciting deep emotional resonance, the influence mechanism follows the classic sequence of “external stimulus (message) → internal affective/cognitive evaluation (brand attitude) → behavioral intention”. However, it is noteworthy that brand attitude did not significantly mediate the path for the sustainable luxury message. This suggests that the influence pathways of different message types may vary: affective messages (e.g., cultural experience) primarily operate through the affective path (influencing attitude), whereas value-laden messages (e.g., sustainability) might involve other, more direct mechanisms (e.g., social norms, personal moral identity) or require specific moderating conditions (e.g., high consumer environmental consciousness) to effectively activate the affective mediating path. This finding deepens our understanding of how IMC messages influence consumer decisions through distinct psychological mechanisms.
Finally, regarding the deeper interpretation of the differences in willingness-to-pay (WTP). Although the ANOVA showed statistically significant differences in WTP across the three groups, the effect size (η2 = 0.236) was substantially lower than those for booking intention (η2 = 0.451) and brand attitude (η2 = 0.422). This indicates that the practical impact strength of message type on WTP is relatively limited during the initial marketing communication stage. A plausible explanation is that Gen Z likely considers luxury travel a high-involvement decision. Upon initial exposure to brand messages, their response is based more on emotional preference and initial attitude (“Do I like it?”) rather than immediately progressing to concrete price evaluation (“How much will I pay?”). Price sensitivity might only be fully activated at a later stage of the decision process (e.g., when comparing specific options). Therefore, while different messages can effectively spark interest and liking, they may be insufficient to cause large differentiations in WTP levels in the short term. This reminds marketers that communication aimed at enhancing premium pricing power may require longer-term value cultivation rather than a single exposure.
5.3. Managerial Implications
First, Core Strategic Shift: From “Property Showcase” to “Destination Cultural Narrative.” The results unequivocally demonstrate the superior efficacy of immersive cultural experience messages in driving positive brand attitude and booking intention among Gen Z. Consequently, brands should pivot their marketing communication focus from traditionally showcasing physical amenities and hardware towards constructing a compelling narrative around the destination’s unique culture and immersive experiences. Specifically, brands should develop serialized high-quality content, such as mini-documentaries, collaborative short films with local artists/artisans, and inviting culturally insightful KOLs for deep immersion experiences. The goal is to position each hotel as a “gateway to the destination’s soul,” rather than merely a luxurious place to stay. This strategy effectively meets Gen Z’s core needs for authenticity, self-extension, and social currency.
Second, Integrating Sustainability Communication: As an Embedded Value, Not a Sales Pitch. While highlighting sustainability alone outperformed basic functional messages, it was less effective than cultural storytelling. This implies that managers should not treat sustainable luxury as an isolated, didactic communication theme, but rather seamlessly weave it into the core cultural narrative. For instance, when promoting local cultural experiences, naturally incorporate how the hotel protects the local environment, supports community artisans, and uses local organic ingredients. Thus, sustainability transforms from an add-on “responsibility claim” into a natural embodiment of the brand’s values—respect for the destination and the pursuit of true luxury. This subtle, “show, don’t tell” approach is more likely to be accepted and valued by Gen Z, who are often skeptical of overt commercial appeals.
Third, Repositioning Flexible Policies: As a Foundational Service Promise to Ensure Decision Confidence. Flexible policies are crucial for Gen Z, but their primary function is akin to a “decision safety net” or “fundamental utility.” This study confirms that promoting them as a central advertising appeal is insufficient to evoke emotional connection and brand preference. Therefore, it is advised to position flexible policies as a clear foundational service promise, prominently displayed at key touchpoints like the official website and app booking engine to reduce perceived risk and eliminate final booking barriers. However, in front-end brand advertising and content marketing, they should cede the spotlight to the more emotionally resonant cultural narrative, playing a role of “enabling” the core story rather than “replacing” it.
5.4. Limitations and Future Research Directions
First, the limitation in sample scope. The respondents were exclusively drawn from the Chinese Gen Z cohort. Although the conclusions are highly valuable for understanding this crucial market, their generalizability may be constrained by cultural context. The values, consumption habits, and understanding of luxury among Gen Z in different cultural settings may vary, thus caution is needed when extending these findings to consumers from other cultural backgrounds.
Second, the simplicity of the experimental environment. The study employed a one-shot online controlled experiment. While this effectively controlled for extraneous variables, the ecological validity is limited compared to the complex process of consumer decision-making in the real world, which involves multi-channel exposure over time. The experiment could not fully simulate the combined effects of dynamic social media influences and peer pressure.
Third, the lack of measurement of long-term behavior. The study measured immediate attitudes and behavioral intentions following message exposure, rather than actual long-term behaviors. The long-term impact of message strategies on brand loyalty remains an open question.
Additionally, as discussed in Chapter 3, there is a consideration regarding the measurement and treatment of Willingness-to-Pay. WTP, inherently an ordinal variable, was treated as continuous in the primary analysis to enhance statistical power. Although supplementary analyses supported the robustness of the findings, this treatment remains a methodological trade-off, suggesting that future research could employ more precise measurement techniques.
Based on these limitations, future research could advance in the following directions:
First, conducting cross-cultural comparative studies. Future research could replicate this experiment in different countries or regions to investigate whether the effects of IMC message strategies on Gen Z vary across cultures, thereby testing the boundary conditions of the current findings and building a luxury hotel marketing theory with a more global perspective.
Second, integrating multimodal data. Future studies could incorporate physiological measures such as eye-tracking and galvanic skin response alongside traditional self-report scales to gain deeper insights into the micro-processes of attention allocation and emotional arousal when consumers process different marketing messages, providing a neuroscientific basis for creative optimization.
Third, exploring message effectiveness in authentic social media contexts. Future research could employ field experiments or big data analytics to deploy different IMC messages on live social media platforms and track actual engagement metrics, thereby assessing the natural propagation efficacy and conversion effectiveness of messages within complex social networks, thus bridging the gap between lab research and marketing practice.