4.1. Measurement Model Assessment
Because the data were collected from the same respondents using a single self-reported questionnaire, common method bias (CMB) was assessed using Harman’s single-factor test (
Podsakoff et al., 2003). The results showed that the first factor accounted for 37.153% of the total variance, which was below the commonly used 50% threshold, suggesting that CMB was unlikely to be a serious concern. In addition, the multicollinearity assessment revealed no serious concerns, as all VIF values ranged from 1.466 to 2.983, remaining well below the conservative cutoff value of 5 (see
Table 3). This indicates that collinearity was unlikely to bias the estimation results (
Hair et al., 2019a;
Kock, 2015).
As reported in
Table 4, all indicator loadings were above the recommended threshold of 0.70 (ranging from 0.746 to 0.883), supporting indicator reliability (
Hair et al., 2022). Internal consistency was also confirmed, with Cronbach’s alpha values between 0.807 and 0.875 and composite reliability values from 0.873 to 0.914, both exceeding the acceptable level of 0.70 (
Hair et al., 2022). Convergent validity was established as AVE values ranging from 0.633 to 0.727, surpassing the minimum requirement of 0.50 (
Fornell & Larcker, 1981). In particular, customer trust (TRU) and hotel reputation (REP) exhibited relatively high AVE values (0.693 and 0.762, respectively), indicating strong explanatory power of their respective indicators in the hotel research context.
Discriminant validity was evaluated using the heterotrait–monotrait ratio (HTMT). As reported in
Table 5, all HTMT coefficients remained below the conservative cutoff of 0.85, with the majority falling under 0.80, thereby satisfying the recommended threshold (
Henseler et al., 2015). Notably, the HTMT values between customer engagement (ENG) and the CSR dimensions were relatively low (ranging from 0.112 to 0.235), suggesting that ENG captures a conceptually distinct construct and can be appropriately modeled as a moderator. In addition, discriminant validity was further verified using the Fornell–Larcker criterion. Specifically, the square root of AVE for each construct exceeded its correlations with other constructs (
Fornell & Larcker, 1981). For example, the square root of AVE for TRU (0.833) and REP (0.873) were both greater than their respective inter-construct correlations, confirming adequate construct separation (
Table 5).
To assess the adequacy of the structural model, a bootstrapping procedure with 5000 resamples was applied to examine the statistical significance of the proposed relationships. Following (
Hair et al., 2016;
Hair et al., 2022;
Hair et al., 2017), model evaluation was based on the coefficients of determination (R
2), predictive relevance (Q
2_predict), and effect size (f
2), rather than relying on a single overall fit index. As shown in
Table 6, the model explained 76.3% of the variance in REP and 62.7% of the variance in TRU, indicating substantial explanatory power for both endogenous constructs. The Q
2_predict values were also positive for REP (0.752) and TRU (0.623), confirming that the model possessed predictive relevance in the present study.
The effect size results in
Table 7 indicate that the individual path effects were generally small, and although the associated f
2 values are modest, the significant paths still make a meaningful incremental contribution to the model’s explanatory power in a multifactor context. For REP, environmental responsibility showed the strongest contribution, whereas TRU, legal responsibility (LEG), economic responsibility (ECO), and customer-oriented responsibility (CUS) also made modest contributions, while community responsibility (COM) had a negligible effect. For TRU, CUS showed the strongest contribution, followed by COM, whereas the remaining CSR dimensions exerted only weak effects. Overall, these findings suggest that the model was suitable for explaining REP and TRU, although the contribution of each individual predictor was limited and should therefore be interpreted with caution (
Hair et al., 2016;
Hair et al., 2022;
Hair et al., 2017).
Collectively, these findings demonstrate that the measurement model satisfies the required criteria for reliability, convergent validity, and discriminant validity, while also showing positive predictive relevance and acceptable but mixed benchmark performance in PLSpredict, thereby providing a solid foundation for subsequent structural model evaluation.
4.2. Structural Model Results
The structural model estimates (see
Table 8 and
Figure 2) show that the majority of the hypothesized relationships are supported. TRU has a positive and statistically significant impact on REP (β = 0.181,
p < 0.001), supporting H1 and suggesting that customer trust functions as an important intervening mechanism in translating CSR perceptions into hotel reputation outcomes. Regarding CSR components, ECO exerts significant positive effects on both REP (β = 0.173,
p < 0.001) and TRU (β = 0.151,
p < 0.001), supporting H2a and H2b. LEG positively influences REP (β = 0.160,
p < 0.001) but does not significantly affect TRU (β = 0.060,
p = 0.058), supporting H3b and not supporting H3a. ENV significantly enhances both REP (β = 0.219,
p < 0.001) and TRU (β = 0.161,
p < 0.001), supporting H4a and H4b. CUS shows the strongest effect on TRU (β = 0.299,
p < 0.001) and also significantly influences REP (β = 0.176,
p < 0.001), supporting H5a and H5b. Finally, COM significantly improves TRU (β = 0.239,
p < 0.001) but does not have a significant impact on REP (β = −0.031,
p = 0.231), supporting H6a and not supporting H6b. Taken together, the direct path results indicate that the five CSR dimensions operate differently across customer trust and hotel reputation. Customer-oriented, environmental, and economic responsibilities positively influence both constructs; legal responsibility mainly affects hotel reputation directly; and community responsibility primarily strengthens customer trust rather than hotel reputation.
Results (
Table 9) showed significant indirect associations for ECO (β = 0.027,
p < 0.001), ENV (β = 0.029,
p < 0.001), CUS (β = 0.054,
p < 0.001), and COM (β = 0.043,
p < 0.001), whereas the indirect association for LEG was not statistically significant (β = 0.011,
p = 0.068). These findings suggest that TRU functioned as an intervening mechanism in the relationships between most CSR dimensions and REP, although the evidence should be interpreted as an indirect association rather than causal mediation. In particular, the coexistence of significant direct and indirect paths for ECO, ENV, and CUS indicates partial mediation by TRU, while the significant indirect path combined with the non-significant direct path for COM indicates full mediation. By contrast, the absence of a significant indirect path for LEG suggests no mediation. Overall, TRU appears to play a differentiated mediating role across CSR dimensions, explaining part of REP’s variance through ECO, ENV, and CUS, fully transmitting the effect of COM, and not mediating the relationship between LEG and REP.
The moderating analysis (
Table 10 and
Figure 3) revealed that customer engagement (ENG) significantly strengthens the relationship between TRU and REP. The direct effect of TRU on REP remained significant (β = 0.181,
p < 0.001), while the interaction term between ENG and TRU was positive and statistically significant (β = 0.080,
p < 0.001), supporting H7. However, the interaction effect was modest, indicating that customer engagement amplified rather than fundamentally altered the customer trust–hotel reputation relationship. The interaction plot further illustrates that the positive effect of customer trust on hotel reputation is stronger at higher levels of customer engagement. This finding suggests that customer engagement adds incremental explanatory value by making the reputational benefits of trust more visible among more involved customers.
4.3. Discussion and Managerial Implications
The findings indicate that CSR influences hotel reputation through a more differentiated mechanism than is often assumed in the literature. Rather than producing a uniform reputational effect, each CSR dimension contributes differently depending on whether customers perceive it as directly service-related, legitimacy-based, or value-driven. This supports the idea that CSR in hospitality is not interpreted as a single abstract construct, but as a set of distinct signals that customers evaluate through their own service experiences and expectations. The results also show that CSR does not always translate directly into hotel reputation: some dimensions operate through customer trust, whereas others influence hotel reputation more directly or not at all. In particular, customer trust emerges as a central relational mechanism linking CSR to hotel reputation, confirming that CSR creates reputational value most effectively when it reduces uncertainty and strengthens confidence in the hotel’s sincerity and reliability. This is consistent with earlier studies by
Fatma et al. (
2018);
Martínez and Del Bosque (
2013), and
O’Connor and Assaker (
2022), who also highlight trust as a key pathway through which CSR influences customer-based outcomes in service contexts.
Among the five dimensions, customer-oriented responsibility emerges as the most influential. This finding suggests that customers respond most strongly to CSR practices directly related to their own interests, such as fair treatment, transparent communication, complaint handling, and service protection. The result is consistent with
Martínez and Del Bosque (
2013),
Chen et al. (
2021), and
O’Connor and Assaker (
2022), who argue that customer-facing CSR is particularly effective because it is visible, personally relevant, and embedded in the service experience. Consequently, customer-oriented responsibility strengthens both customer trust and hotel reputation, highlighting its strategic importance in hotel reputation management.
Economic responsibility also contributes positively to both customer trust and hotel reputation. This finding is consistent with
Park et al. (
2014),
H. L. Kim et al. (
2017) and (
Latif et al., 2020), who reported that economically responsible practices enhance perceptions of competence, reliability, and reputation. In the hotel context, stable operations, efficient resource management, and consistent service quality appear to function primarily as signals of organizational competence. This result further supports the argument of
González-Rodríguez et al. (
2021) that the effectiveness of CSR dimensions depends on their visibility and relevance to customers.
Environmental responsibility likewise exerts positive effects on both customer trust and hotel reputation. This finding aligns with
González-Rodríguez et al. (
2021) and
Ali et al. (
2023), who emphasize that environmental CSR is most effective when it is visible and embedded in hotel operations. Practices such as energy conservation, waste reduction, and eco-friendly amenities signal responsibility and long-term commitment, thereby enhancing both trust and reputation. These results reinforce the growing importance of environmental responsibility as a strategic component of sustainable hotel development.
The pattern for legal responsibility is more nuanced. It contributes to hotel reputation, but it does not meaningfully strengthen customer trust. This divergence is theoretically important. It suggests that legal compliance functions primarily as a baseline expectation rather than as a distinctive CSR signal. Customers appear to assume that hotels should obey laws and regulations, so compliance may protect hotel reputation, but it does not necessarily create additional customer trust. This finding differs from
Park et al. (
2014) and
H. L. Kim et al. (
2017), who reported stronger trust-related effects of legal responsibility in service settings. One possible explanation is contextual: in the Mekong Delta hospitality market, customers may view compliance as a minimum requirement rather than as evidence of moral excellence or relational commitment. Therefore, legal responsibility should be understood as a hygiene factor. It is necessary for legitimacy, but it is unlikely to generate a strong reputational advantage on its own.
Community responsibility presents another interesting pattern. It contributes to customer trust but does not directly translate into hotel reputation. This means that community-oriented CSR works more as a moral signal than as an immediate reputational driver. Supporting local development, charitable initiatives, or social causes may increase customers’ goodwill toward the hotel, but these activities may not be sufficiently visible or personally relevant to generate hotel reputation unless they are clearly communicated and connected to the guest experience. This finding is broadly aligned with
Chen et al. (
2021) and
Ali et al. (
2023), who note that CSR initiatives need authenticity and salience to become reputational assets. In practical terms, community responsibility should not be hidden behind internal activities. Hotels need to make their social value understandable to customers; otherwise, its impact remains indirect.
The mediation results further clarify the mechanism. Customer trust serves as an important bridge through which economic, environmental, customer-oriented, and community responsibility influence hotel reputation, but not legal responsibility. More precisely, customer trust operates as a partial mediator for economic, environmental, and customer-oriented responsibility, a full mediator for community responsibility, and not as a mediator for legal responsibility. This suggests that customer trust is especially important when CSR is seen as voluntary, authentic, and customer-relevant. Where CSR is interpreted as obligatory, as in the case of legal responsibility, it may support hotel reputation directly but does not necessarily deepen customer trust. This distinction adds theoretical clarity because it shows that CSR dimensions differ not only in strength but also in the psychological route through which they operate. Some CSR practices build hotel reputation because they are seen and experienced directly; others build hotel reputation because they first strengthen confidence in the hotel’s motives and reliability.
The moderating effect of customer engagement adds another layer of insight. The relationship between customer trust and hotel reputation becomes stronger when customers are more engaged. This confirms the view of
So et al. (
2014),
Hollebeek and Macky (
2019), and
Rather et al. (
2023) that customer engagement amplifies the value of relational outcomes. However, the interaction effect is statistically significant but modest, indicating that customer engagement strengthens rather than fundamentally reshapes the customer trust–hotel reputation relationship. This study therefore extends prior work by showing that customer engagement is not merely an outcome of good relationships; it is also a condition that strengthens how customer trust is converted into reputational capital. Engaged customers pay more attention to hotel behavior, process CSR information more deeply, and are more likely to translate customer trust into word of mouth, loyalty, and favorable hotel reputation. This means that customer trust alone is not enough. Hotels must also cultivate active customer participation if they want customer trust to become a durable reputational asset.
From a managerial perspective, the results suggest that hotel managers should adopt a selective and differentiated CSR strategy rather than a general one. First, they should prioritize customer-oriented responsibility because it delivers the strongest reputational return. This means emphasizing fairness, transparency, responsive complaint handling, and service recovery. Second, environmental responsibility should be made visible through concrete practices that guests can observe during their stay. Third, economic responsibility should be presented as evidence of professionalism, stability, and consistency. Fourth, legal responsibility must be maintained rigorously, but managers should recognize that compliance alone will not generate a strong trust advantage. Fifth, community-oriented responsibility should be communicated in ways that make its social value meaningful to guests. For hotels in the Mekong Delta, this means linking CSR more concretely to local suppliers, river- and orchard-based tourism experiences, cultural preservation, waste reduction, energy-saving practices, and credible communication through the channels that guests actually use. CSR messages should therefore be specific, visible, and authentic rather than general and declarative. Finally, hotels should actively build customer engagement through personalized service, digital interaction, loyalty communities, feedback systems, and co-created experiences, because customer engagement increases the reputational payoff of customer trust.