Hotels classify CSR practices into many verticals, such as philanthropic, legal-reactive, and active approaches, with the primary objective of enhancing economic performance. This classification helps us understand the CSR management model that benefits the industry ( Peña-Miranda et al., 2021). Many organizations have integrated CSR into their operational strategies and promoted company sustainability and stakeholder engagement ( Bohdanowicz & Zientara, 2008). However, CSR activities related to community engagement and environmental issues do not always significantly impact reputation, indicating that integrating CSR with organizational culture is crucial for performance outcomes ( González-Rodríguez et al., 2019). Recent shifts towards communication and reporting suggest a growing emphasis on transparency and accountability in CSR practices ( Moyeen & Mehjabeen, 2024). CSR initiatives have been widely recognized as tools for enhancing business performance and stakeholder perceptions ( Sultan et al., 2024). Many studies confirm that CSR positively influences customer loyalty through different mediating factors such as image, trust, customer satisfaction, and social platform use ( Liu et al., 2019; Su et al., 2014; Mohammed & Al-Swidi, 2019; Islam et al., 2022). The above arguments support the development of H1 i.e, H1

Customer perception significantly affects Customer loyalty.

CSR not only affects hotel reputation but also customers’ emotions ( Su et al., 2014). A positive hotel reputation further impacts customer loyalty ( Kandampully & Suhartanto, 2000). For hotels, it is essential to comprehend the link between customer satisfaction and their intention to return ( Arora & Singer, 2006). Furthermore, the quality of service provided by a hotel is equally significant, as it greatly influences customer satisfaction and loyalty ( Akbar et al., 2010). CSR initiatives enhance a hotel’s reputation, which in turn boosts trust, satisfaction, and loyalty towards the company.

Such initiatives further build customer satisfaction, which serves as a mediator between CSR and CL ( Lee, 2018). Additionally, a hotel’s ethical perception can enhance customer identification with the brand, leading to emotional commitment and brand trust, thus ensuring long-term loyalty ( Fatma & Rahman, 2017). Moreover, customer experiences in full-service hotels, which encompass functional, emotional, and social dimensions, significantly impact brand trust, leading to CL ( Guan et al., 2021). The dynamic interplay between quality of service, hotel image, and customer satisfaction plays a fundamental role in establishing customer loyalty. Each of these elements contributes uniquely yet interdependently to shaping consumers’ overall perception of a brand or organization. High service quality enhances perceived value, and a strong corporate image fosters trust and credibility. Collectively, they influence consumer attitudes and behaviors, thereby serving as essential drivers in the creation and maintenance of a stable, committed, and loyal customer base ( Cheng, 2014). The above arguments advocate the development of H2, that is H2

Customer perception of CSR significantly affects hotel reputation.

CSR activities contribute not only to social value but also to strengthening hotel reputation and enhancing customer loyalty. It positively affects hotel reputation, which subsequently increases customer loyalty through improved trust. Reputation is a trust-building factor that deals with intangible services. However, limited research has explored how reputation acts as a psychological factor that transforms CSR perceptions into loyalty behavior ( He & Li, 2011; Martínez & Rodríguez del Bosque, 2013). Stakeholder theory suggests that customers tend to reward companies that reflect their social values, although these rewards are typically not immediate and instead manifest through constructs such as reputation, which accumulate perceptions over time ( Fombrun & Shanley, 1990; Freeman, 1984). When hotels implement CSR activities, such as promoting environmental sustainability, engaging with the community, or adopting ethical labor practices, they send positive messages to the public, contributing to the development of a strong, positive corporate reputation ( He & Li, 2011). This reputation, in turn, enhances customer trust, brand admiration, and psychological closeness, all of which are precursors to loyalty behaviors such as repeat visits and advocacy ( Walsh, Beatty, & Shiu, 2009). CSR activities have significantly boosted the public image of hotels, and this enhanced reputation directly influences customers’ willingness to recommend and return ( Lee et al., 2013). This suggests that hotel reputation works as a value transformation mechanism, where a brand’s ethical and social efforts are converted into loyalty outcomes through positive stakeholder perception. Kang, Lee, and Huh (2010) also highlighted that while CSR directly impacts performance, its effect is magnified when the hotel is seen as reputable and credible. In other words, CSR alone may not be sufficient to secure customer loyalty unless customers perceive it as a reputation-enhancing activity ( Pérez & Rodríguez del Bosque, 2015). Customers often cannot directly assess the authenticity of CSR actions; therefore, reputation serves as a cognitive shortcut that helps them infer the long-term intentions and reliability of the hotel brand ( Groza et al., 2011). When both CSR and reputation are viewed positively, customers are more likely to develop affective commitments, making them less sensitive to price and more emotionally invested in the brand. Thus, hotel reputation is not just a by-product of CSR but a strategic asset that mediates trust-building and relational value derived from socially responsible practices. Thus, the arguments presented above provide a basis for formulating H3 and H4, respectively. H3

Hotel reputation significantly affects customer loyalty.

The conceptual framework for the present study is depicted in Figure 1.

To empirically examine the hypothesized relationships among the constructs CSR, HR, and CL, a system of linear equations is proposed. HR = β 1 · CSR + ε 1 (1) CL = β 2 · CSR + β 3 · HR + ε 2 (2)

Whereas: •

CSR = Customer perception of corporate social responsibility

The first equation expresses HR as a function of CSR. This implies that hotel’s CSR has a direct influence on how the hotel’s reputation is perceived, where β ₁ represents the path coefficient, and ε ₁ accounts for unexplained variance or error. The second equation outlines CL as influenced by both CSR and HR. Here, CL is shaped not only directly by CSR (coefficient β ₂) but also indirectly through the mediating effect of Hotel Reputation (coefficient β ₃). ε ₂ represents the residual error in predicting customer loyalty.

Figure 3 illustrates the outer and inner loadings of the measurement model. To assess the internal consistency of the constructs, multiple reliability indicators were employed, including Cronbach’s alpha, rho_A, and composite reliability (rho_ C) ( Table 2). The results revealed that all factor loadings were above the acceptable threshold of 0.70, indicating strong reliability. Furthermore, Cronbach’s alpha values exceeded 0.70 for all constructs, indicating acceptable internal consistency. The rho_A values, which provide a more accurate estimate of construct reliability in certain contexts, fell between Cronbach’s alpha and rho_ C values and consistently remained above 0.70. This alignment among the reliability measures confirms that the constructs exhibit strong internal reliability and consistency. To establish convergent validity, the average variance extract was used. As presented in Table 2, the Average Variance Extracted (AVE) values for all constructs were greater than the recommended threshold of 0.50. This indicates that the measurement model exhibited acceptable convergent validity, confirming that the items intended to measure each construct shared a sufficient proportion of common variance. In addition to convergent validity, discriminant validity was assessed using several approaches. These included examining the cross-loadings of each indicator, applying the Fornell–Larcker criterion, and calculating the hetero trait–mono trait ratio of correlations (HTMT). The results from all three methods demonstrated that the constructs CSR, HR, and CL are empirically distinct from one another, thereby supporting the establishment of discriminant validity. The outcomes of all three tests are detailed in Table 3, Table 4, and Table 5. Multicollinearity for all factors was examined using VIF ( Table 2). The VIF values for all factors were below 5, suggesting no multicollinearity concerns.

The hypothesized relationships among the three core constructs–CSR, HR, and CL–are shown in Figure 3. The model was tested using path analysis with the PLS-SEM approach.

H1 explores the link between CSR and CL (CSR → CL). The analysis reveals that CSR significantly impacts CL (β = 0.357, t = 8.456, p < 0.01), thereby validating H1. H2 considers whether CSR is significantly linked to HR (CSR → HR). The findings confirm that CSR significantly impacts HR (β = 0.562, t = 16.104, p < 0.01), confirming H2. H3 investigates the significant association between HR and CL (HR → CL). The results showed that HR significantly impacted CL (β = 0.388, t = 8.823, p < 0.01), leading to the acceptance of H3. The outcomes are presented in Table 6.

H4 examined whether HR acts as a mediator between CSR and CL. Specifically, it examined whether the influence of CSR on CL is transmitted through HR, forming an indirect path (CSR → HR → CL). The results revealed a statistically significant indirect effect (β = 0.218, t = 7.807, p < 0.05). This indicates that CSR positively influences CL through its effect on HR. Path analysis showed that in the presence of HR as a mediator, the direct effect of CSR on CL remained statistically significant (β = 0.357, t = 8.456, p < 0.01). As both effects were significant, this suggests that HR partially mediates the CSR-CL relationship ( Table 7).

Table 8 presents the findings related to the model’s explanatory and predictive strengths. The R ² value for CL was 0.435, indicating that the combined effects of CSR and Hotel Reputation (HR) accounted for 43.5% of the variance in CL. The results show a moderate level of explanatory power. Hence, the model effectively captures the key determinants of customer loyalty in the hotel industry.

The results also revealed that both CSR and HR exert a moderate effect on CL (f ² > 0.15), while CSR alone demonstrated a large effect size (f ² > 0.35), underscoring its substantial role in shaping customer loyalty. The Q ² value was assessed using a blindfolding procedure. The Q ² values for HR and CSR were 0.327 and 0.313, respectively, both surpassing the recommended threshold of 0.15, as suggested by Hair et al. (2013). Hence, the model possesses medium predictive relevance, indicating its capability to predict endogenous constructs with reasonable accuracy.