Green Behavioural Intention and Behaviour of Hotel Employees: Mediation Roles of Customers, Coworkers, Supervisors, and Corporate Attitudes

Abstract

This study investigates the mechanisms through which employees’ green behavioural intention (GBI) translates into employee in-role green behaviour (EGB-IR) in the hospitality sector. Drawing on the Theory of Planned Behaviour (TPB) and the Value-Belief Norm (VBN) theory, we propose an extended model that incorporates four contextual mediators —customers’ environmental attitudes (CEAs), coworkers’ green work climate (CGWC), supervisors’ commitment to the environment (SCE), and environmental organisational policy (EOP)—to explain how the relationship between intentions and behaviours is mediated by these social and organisational factors. Data were collected through a self-administered survey of 497 employees from hotels located in the Canary Islands, a recognised coastal tourism destination facing relevant environmental challenges. The proposed relationships were tested using Partial Least Squares Structural Equation Modelling (PLS-SEM). Results confirm that GBI significantly predicts in-role green behaviour (EGB-IR), and all four contextual factors partially mediate this relationship. Among them, EOP and CGWC emerged as the most influential mediators. These findings underscore the importance of aligning individual motivation with supportive workplace environments to foster consistent sustainable practices in hospitality organisations. This research contributes to the growing literature on green behaviours in hospitality by empirically validating a multilevel TPB-VBN-theory-based framework and highlighting key points for hotel managers aiming to strengthen their environmental commitment through employee engagement.

1. Introduction

To what extent do hotel employees’ intentions to act sustainably at work translate into actual green behaviours, and which organisational and social factors help bridge this intention–behaviour gap? As the global tourism industry faces increasing environmental scrutiny, the role of employee behaviour in advancing sustainability goals has gained growing academic and managerial attention [1,2]. In the hospitality sector, sustainability has shifted from a reputational and positioning concern to a strategic aspect to meet stakeholder expectations, regulatory frameworks, and market pressures [3,4]. However, many environmental management systems fail to achieve tangible outcomes, not necessarily due to strategic issues, but often due to limited behavioural engagement from frontline employees [5,6].

This behavioural dimension is critical in hotels, where service delivery, guest interaction, and operational processes are inherently human-driven. Scholars increasingly emphasise that employees’ daily green behaviours constitute the “last mile” of corporate environmental strategies, without which sustainability commitments risk remaining rhetorical [7,8]. Yet, the psychological, social, and organisational mechanisms that facilitate or hinder the translation of green behavioural intention (GBI) into actual employee in-role green behaviour (EGB-IR) remain underexplored.

The Theory of Planned Behaviour (TPB) [9] has provided a foundational framework for understanding the concept of behavioural intentions and their translation into action. According to TPB, attitudes, subjective norms, and perceived behavioural control shape intention, which in turn predicts behaviour. However, growing empirical evidence suggests that the link between intention and behaviour is frequently incomplete or inconsistent, particularly in organisational contexts where social norms, leadership, and structural support critically influence behavioural outcomes [10,11,12]. This highlights a crucial gap in the implementation of sustainability strategies, calling for more comprehensive theoretical frameworks that go beyond individual intention to account for social and moral drivers of action.

Recent research has therefore called for expanded TPB-based models that incorporate relational, normative, and institutional mechanisms [2,12]. In this context, the Value-Belief Norm (VBN) theory [13] highlights that personal values, ecological beliefs, and moral obligations also contribute to shaping sustainable behaviours, suggesting that behavioural engagement is driven not only by planned cognition, but by deeper normative commitments. By integrating both TPB and VBN, this study adopts a comprehensive perspective on the complex framework in which green behaviours appear in the workplace. However, existing research often focuses on either individual-level predictors (e.g., environmental values, attitudes) or organisational strategies, without fully integrating these dimensions within a comprehensive behavioural model. Moreover, although the TPB and VBN frameworks have been widely applied in environmental psychology, their combined use in workplace sustainability contexts remains limited.

This study addresses the following research question: to what extent does green behavioural intention (GBI) predict in-role green behaviour (EGB-IR) among hotel employees, and how do social and organisational contextual factors mediate this relationship? Drawing on an integrated framework combining the Theory of Planned Behaviour (TPB) and the Value-Belief Norm (VBN) theory, the study proposes a multilevel mediation model. Specifically, it examines the mediating role of four contextual factors in the relationship between employees’ green behavioural intention and their actual sustainable practices at work: (1) customers’ pro-environmental attitudes, (2) coworkers’ green work climate, (3) supervisors’ commitment to the environment, and (4) perceived environmental organisational policy. By empirically validating this extended model in a sample of 497 hotel employees in the Canary Islands, the study offers a more holistic understanding of sustainable behaviour in hospitality contexts, contributing to both theoretical development and managerial practice.

Understanding these dynamics is crucial for informing effective, employee-centred sustainability strategies [14]. By investigating how employees’ intentions, social perceptions, and organisational structures interact in hotels, this study seeks to bridge a critical gap in behaviourally grounded hospitality research. It also may have relevant implications for managers striving to translate sustainability strategies into authentic operational practices.

2. Theoretical Framework and Hypotheses Development

2.1. Theory of Planned Behaviour and Value-Belief Norm Theory

The TPB [9] has long been established as a foundational framework for understanding human behavioural processes, particularly in contexts where personal intentions serve as precursors to action. According to TPB, behaviour is determined primarily by behavioural intention, itself shaped by attitudes, subjective norms, and perceived behavioural control. Its application to sustainability research has been extensive, notably in explaining recycling, energy conservation, and sustainable consumption behaviours [12].

In the hospitality sector, TPB has gained prominence in analysing the translation of employees’ intentions into environmentally responsible practices at work [6,10,15]. Yet, the pathway from intention to behaviour is not always guaranteed. Webb and Sheeran [11], in a meta-analysis, demonstrated that intentions only partially predict behaviours, revealing the critical influence of contextual and organisational factors in enabling or constraining green workplace behaviours.

Recent studies have criticised TPB for underestimating the broader social, normative, and organisational structures in which behaviours may happen [2]. Research calls for enriched frameworks that integrate relational and institutional mechanisms alongside individual cognition has emerged lately [16,17]. In this context, the VBN theory is incorporated [13], which posits that behaviour is not merely a function of planned intention, but is also driven by internalised values, ecological worldviews, and personal moral obligations.

While TPB provides a cognitive–rational account of how attitudes, perceived control, and social norms influence behavioural intention, it has been criticised for overlooking deeper moral motivations and value-based drivers [2,17]. In contrast, the VBN theory [13] posits that environmental behaviours also stem from internalised ecological values and a sense of moral obligation.

Integrating TPB and VBN enables a more comprehensive understanding of how green behaviours emerge in the workplace. While the TPB provides a cognitive and rational foundation—emphasising intention as a key predictor of behaviour shaped by attitudes, subjective norms, and perceived behavioural control—the VBN introduces moral and normative drivers that explain why some individuals may act beyond strategic calculation (

Table 1. Integration of TPB and VBN models.

Dimension	Theory of Planned Behaviour (TPB) [9]	Value-Belief Norm Theory (VBN) [13]
Main Assumption	Behaviour is guided by intention, based on reasoned decision-making	Behaviour is driven by internal values and moral norms
Core Constructs	Attitudes, Subjective Norms, Perceived Behavioural Control, Intention	Values, Ecological Beliefs, Personal Norms
Behavioural Mechanism	Intention → Behaviour (moderated by control and norms)	Norm Activation → Moral Obligation → Behaviour
Focus	Cognitive and social factors influencing intention	Internalised norms and value systems
Contribution to this study	Explains how intention forms based on perceived control and expectations	Explains why intention may activate or fail based on moral obligation
Integration in this model	GBI as central predictor; context perceptions as extensions of norms	Context perceptions as norm-activating stimuli reinforcing intention

). Specifically, green behavioural intention (GBI) in our model is grounded in the TPB’s emphasis on planned action, while also reflecting internalised environmental values and personal norms as suggested by the VBN framework. Likewise, in-role green behaviour (EGB-IR) is not seen solely as the outcome of rational planning but also as the enactment of ecological self-identity and moral obligation. Among the mediators, constructs such as customers’ environmental attitudes (CEAs) and supervisors’ environmental commitment (SCE) reflect both normative pressures (TPB) and the activation of personal norms (VBN) through observation and social influence.

This enriched approach is particularly relevant in service industries such as hospitality, where employee behaviour is simultaneously influenced by personal attitudes and external expectations from customers, colleagues, supervisors, and organisational policies [18,19]. Thus, this study proposes a conceptual model in which employees’ GBI predicts their EGB-IR, while four contextual factors act as mediators: CEA, CGWC, SCE, and the EOP.

2.2. Employees Green Behavioural Intention and Green Behaviour at Work

GBI refers to the degree to which individuals plan or intend to engage in environmentally responsible behaviours in the workplace. Grounded in the TPB [9], intention is widely assumed as a determinant of actual behaviour. This relationship has been confirmed in a wide range of areas, including health, consumption, and environmental actions [11,12].

In the hospitality context, several studies have supported the predictive power of GBI for pro-environmental workplace behaviours. For instance, Greaves et al. [6] found that employees with stronger intentions were more likely to implement environmentally friendly practices in hotels. Similarly, Okumus et al. [10] and Viglia and Acuti [15] confirmed that intention significantly predicts green behaviours in service companies, although the magnitude of this effect may vary depending on the contextual conditions.

In connection, while GBI is a strong predictor, recent literature has recognised that intention alone may not be sufficient to ensure consistent green behaviour at work [8,12,20]. Environmental behaviour is often shaped not only by individual motivation, but also by the normative, relational, and organisational environments in which employees operate [3,21]. Particularly in labour-intensive sectors like hospitality, the alignment between intention and actual behaviour is influenced by factors such as peer support, managerial commitment, and policy coherence.

A growing body of research highlights that the relationship between green behavioural intention (GBI) and actual green behaviour is not always straightforward. Several studies report inconsistencies, showing that strong pro-environmental intentions do not always translate into workplace actions, especially when employees face contextual or structural constraints [22,23,24]. For example, recent studies show that contextual factors such as green leadership, perceived coworker support, and environmental policies significantly shape how behavioural intentions manifest in action [25]. Additionally, the presence of a supportive green psychological climate has been shown to moderate the relationship between intention and behaviour, facilitating behavioural consistency when environmental values are embedded in the organisational environment [26]. These findings suggest that GBI may interact with normative and organisational cues, which can act as either enablers or inhibitors of behaviour.

Supportive organisational conditions—such as green human resource management (GHRM) practices, ethical leadership, or clear sustainability policies—are often necessary to enable the effective enactment of green intentions [27,28]. Further evidence shows that environmental knowledge sharing and application within the organisation enhance employees’ ability to translate green intentions into action, especially when supported by perceived organisational backing. Similarly, green servant leadership has been associated with higher green self-efficacy and engagement, both of which reinforce the enactment of behavioural intentions [29]. Additionally, green organisational climates and peer norms—such as coworkers’ green practices and supervisors’ environmental commitment—have been empirically linked to stronger behavioural consistency among employees with a high GBI [30,31].

Furthermore, perceptions of organisational support and psychological contract fulfilment can significantly shape employees’ willingness to act on their ecological values [26,28]. Therefore, recent research underscores the relevance of integrating social and organisational variables when analysing the GBI-EGB relationship, particularly in service industries where culture, leadership, and peer dynamics create the conditions under which intentions are either enabled or suppressed [32].

Therefore, although the direct link between GBI and EGB-IR has been previously established, this study examines the relationship in the specific context of hotels. The aim is to assess whether intention remains a robust predictor in environments where sustainability practices are formally recognised and to what extent contextual mediators may influence this relationship, aligning with recent calls to enrich the TPB framework through social and institutional lenses [2]. Consequently, we hypothesise the following:

H1. 

Employees’ green behavioural intention (GBI) is positively associated with their in-role green behaviour (EGB-IR).

In addition to predicting EGB-IR directly, employees’ GBI is also expected to influence key contextual perceptions. In line with the TPB [9] and the VBN theory [13], individuals’ behavioural intentions can shape how they interpret social and organisational environments. Employees with stronger green intentions are more likely to perceive greater support from supervisors, alignment with coworkers’ sustainable practices, sensitivity to customers’ environmental expectations, and a higher congruence with organisational sustainability policies. Accordingly, the following hypotheses are proposed:

H2. 

Employees’ green behavioural intention (GBI) is positively associated with supervisors’ commitment to the environment (SCE).

H3. 

Employees’ green behavioural intention (GBI) is positively associated with coworkers’ green work climate (CGWC).

H4. 

Employees’ green behavioural intention (GBI) is positively associated with customers’ environmental attitudes (CEAs).

H5. 

Employees’ green behavioural intention (GBI) is positively associated with perceptions of environmental organisational policy (EOP).

2.3. Mediators

Drawing upon the TPB and the VBN theory, this study examines four contextual factors that may mediate the relationship between employees’ green behavioural intention (GBI) and their in-role green behaviour (EGB-IR)—modelled as parallel mediators in the PLS-SEM framework, meaning that they exert their influence independently and simultaneously—supervisors’ commitment to the environment, coworkers’ green work climate, customers’ environmental attitudes, and organisational environmental policy. These specific mediators were selected based on their relevance to service-sector workplaces, where sustainability behaviours are shaped by dynamic interactions among individual values, social norms, and institutional frameworks. The TPB emphasises the role of subjective norms and perceived behavioural control [9], while the VBN theory highlights the activation of personal moral norms [13]. The chosen mediators capture key normative and structural forces operating across different relational levels—vertical (supervisors), horizontal (coworkers), external (customers), and systemic (organisational policy)—that have been repeatedly identified as influential in prior research [3,8,22,33]. By focusing on these four dimensions, the model reflects a multi-level understanding of how workplace contexts can either facilitate or constrain the realisation of employees’ pro-environmental intentions.

2.3.1. Supervisors’ Commitment to the Environment

Supervisors play a central role in shaping employees’ behaviour through leadership, reinforcement, and the allocation of resources and recognition. In sustainability-oriented workplaces, the perceived commitment of supervisors to environmental values can significantly influence whether employees feel supported and encouraged to act in environmentally responsible ways [33,34].

When employees perceive that their supervisors genuinely value sustainability, they are more likely to engage in voluntary pro-environmental behaviours [35]. In this regard, supervisors function as key players between individual intention and actual behaviour, translating broad environmental goals into daily routines and expectations [36,37].

Empirical studies have confirmed that green transformational leadership and supervisors’ environmental support positively affect employees’ motivation and green workplace behaviour [38]. Moreover, Erdogan et al. [33] found that when supervisors model pro-environmental conduct and explicitly support sustainability initiatives, employees are more likely to overcome barriers to implementing green practices, even when such behaviours go beyond formal job expectations.

In contexts such as hospitality, where frontline staff operate under considerable time pressure and customer expectations, perceived supervisor attitudes can act as a permission structure, reducing the perceived risk of deviating from standard procedures in order to act sustainably [16]. Based on this, we hypothesise the following:

H6. 

Supervisors’ commitment to the environment (SCE) is positively associated with employees’ in-role green behaviour (EGB-IR).

Moreover, supervisory environmental commitment may mediate the relationship between employees’ green behavioural intentions and their sustainable behaviour. Supervisors may reinforce normative pressures and activate personal norms that facilitate the translation of intention into behaviour. Thus,

H7. 

Supervisors’ commitment to the environment (SCE) mediates the relationship between green behavioural intention (GBI) and employees’ in-role green behaviour (EGB-IR).

2.3.2. Coworkers’ Green Work Climate

The immediate work environment plays a fundamental role in shaping employees’ behaviours, especially in sectors characterised by high levels of social interaction and interdependence, such as hospitality. Within this environment, coworkers’ shared attitudes and behaviours regarding environmental sustainability form what is referred to as a green work climate [22]. This concept reflects employees’ perceptions of the extent to which their colleagues value and enact environmentally responsible practices.

Coworkers could become key referents for behavioural alignment, particularly in ambiguous or demanding service environments. When pro-environmental behaviours are visibly and consistently performed by peers, they are more likely to be adopted by others, not only through imitation but also through a process of norm reinforcement [18,36].

Previous research has demonstrated that coworker support and team-level sustainability norms positively influence individual engagement in green behaviours at work [36,39]. Moreover, the presence of a green work climate enhances the translation of environmental values and intentions into daily practices, acting as a social amplifier of pro-environmental commitment. In contrast, when coworkers exhibit indifference or resistance to sustainability, even motivated individuals may experience pressure to adapt to the environment, potentially reducing the behavioural impact of their intentions [18,36].

In the context of sustainable hotels, where sustainability is embedded into operational practices, the reinforcement (or lack thereof) of pro-environmental norms among peers can significantly affect the consistency and visibility of green actions. Therefore, we propose the following:

H8. 

Coworkers’ green work climate (CGWC) is positively associated with employees’ in-role green behaviour (EGB-IR).

Moreover, coworkers’ collective environmental engagement is expected to mediate the relationship between employees’ green behavioural intentions and their sustainable behaviours. A supportive peer environment amplifies the translation of personal intention into consistent green practices. Therefore,

H9. 

Coworkers’ green work climate (CGWC) mediates the relationship between green behavioural intention (GBI) and employees’ in-role green behaviour (EGB-IR).

2.3.3. Customers’ Environmental Attitudes

In hospitality, customers play a key role, whose behaviours and expectations influence employee actions. While the TPB traditionally focuses on peers and authority figures as normative sources, recent studies have extended this scope by identifying customers as influential external actors, especially in service environments where guest feedback and evaluation strongly shape workplace behaviour [8,21,40]. From the perspective of the VBN theory, customers’ visible environmental attitudes may also act as external stimuli that activate personal moral norms within employees. Observing clients’ commitment to sustainability can lead employees to experience a heightened sense of personal obligation to align their actions with environmental values [22,37].

Moreover, service norm theory suggests that customers transmit implicit expectations that influence frontline staff behaviour, acting as informal sources of normative pressure [40]. When employees perceive pro-environmental concern from guests, these perceptions may reinforce their own green behavioural intentions and strengthen behavioural consistency. Conversely, when apathy or scepticism is perceived, sustainability-related efforts may be deprioritised, even in the presence of personal motivation.

Empirical studies in hospitality and tourism have shown that customers’ values and expectations act as powerful contextual factors, influencing employee engagement and sustainability performance [16]. This is particularly relevant in front-office roles or guest-facing departments, where interactions are frequent and performance is evaluated in part through customer satisfaction. In these contexts, employee behaviour is often guided by implicit social expectations, reinforcing the need to consider clients’ perceived sustainability attitudes as a meaningful mediator. Thus, clients’ attitudes are not only exogenous perceptions but can become internalised workplace norms that condition behavioural outcomes. Accordingly, this study hypothesises that

H10. 

Customers’ environmental attitudes (CEA) are positively associated with employees’ in-role green behaviour (EGB-IR).

Furthermore, customers’ environmental attitudes are expected to mediate the link between employees’ green behavioural intention and their in-role green behaviours, acting as a reinforcing contextual factor. Thus, we propose the following:

H11. 

Customers’ environmental attitudes (CEA) mediate the relationship between green behavioural intention (GBI) and employees’ in-role green behaviour (EGB-IR).

2.3.4. Environmental Organisational Policy

Actual employees’ behaviour may be also shaped by the formal structures, policies, strategies, and environmental management systems implemented by their organisations. The presence of a clear, visible, and operational environmental organisational policy (EOP) signals institutional commitment to sustainability, reinforcing employees’ pro-environmental intentions and legitimising their behaviour at work [8].

Organisational policies reflect the top management’s strategic priorities and provide both symbolic and instrumental support for desired behaviours. EOP constitutes formal rules and normative expectations that guide employee behaviours. In the context of sustainability, environmental policies serve to embed ecological values within operational routines, performance metrics, and resource allocation [41,42].

Empirical research has demonstrated that employees are more likely to engage in green behaviours when they perceive that their organisation has implemented structured environmental initiatives, such as sustainability reporting, environmental training, or green purchasing protocols [8]. These policies reduce ambiguity, signal consistency, and provide legitimacy for environmentally responsible actions, even when these actions require additional effort or involve deviating from established routines [3,14]. Therefore, it is hypothesised that

H12. 

Environmental organisational policy (EOP) is positively associated with employees’ in-role green behaviour (EGB-IR).

Moreover, by enhancing perceived behavioural control and activating personal moral norms, environmental policies are expected to mediate the relationship between intention and behaviour. Thus, we hypothesise the following:

H13. 

Environmental organisational policy (EOP) mediates the relationship between green behavioural intention (GBI) and employees’ in-role green behaviour (EGB-IR).

2.4. Conceptual Model Proposal

Based on an extensive review of the literature, a conceptual model has been developed. This model posits that GBI positively predicts their sustainable behaviour in the workplace. Furthermore, it is proposed that the customers’ environmental attitudes (CEAs), coworkers’ green work climate (CGWC), and supervisors’ commitment to the environment (SCE), as well as the organisational environmental policy (EOP), act as mediators in this relationship. These mediating factors not only influence the strength of the relationship between GBI and sustainable behaviour but also contribute directly to shaping employees’ engagement in green practices.

Building on the extended framework of the TPB, incorporating the VBN theory [9,13], this study acknowledges that sustainable behaviours are shaped not only by individual intentions but also by social, normative, and organisational factors. By integrating both intentional and normative components, the proposed model offers a more comprehensive understanding of how green behaviours emerge in hospitality work contexts (Figure 1).

3. Methodology

3.1. Data Collection

Data were collected through a self-administered online survey conducted among hotel employees in the Canary Islands, one of Europe’s leading mass tourism destinations. The questionnaire was distributed with the support of the regional hoteliers’ association via email to hotel managers, who facilitated internal employee participation. The instrument was developed in English, translated into Spanish by bilingual experts, and back-translated to ensure semantic accuracy. Prior to the main data collection, a pilot test with ten employees was conducted to ensure the clarity and contextual relevance of the items. Minor linguistic and formatting adjustments were made based on participant feedback. The survey was open from January to June 2024.

A total of 530 responses were initially collected, but following a rigorous three-step data quality control process—including checks for the response time, attention filters, and detection of straightliners—497 valid responses were retained for analysis. This final dataset was used to test the model and hypotheses via Partial Least Squares Structural Equation Modelling (PLS-SEM). Hotel inclusion was voluntary from a broad range of properties across the archipelago.

3.2. Measures

All constructs were modelled reflectively and measured using previously validated scales adapted to the hotel sector context (Appendix A). This choice aligns with the theoretical nature of the variables, which represent underlying latent traits (e.g., intention, perception) believed to cause the observed responses. Following Jarvis et al. [43], reflective models are appropriate when items are manifestations of the same construct and are expected to covary. Additionally, confirmatory factor analysis showed adequate internal consistency, supporting the reflective specification (Table 4). Theoretically, all constructs represent unidimensional latent traits expected to cause observed item responses [43] (Jarvis et al., 2003). Empirically, all items showed high and significant loadings (≥0.70), the composite reliability (CR) values exceeded 0.80, and average variance extracted (AVE) values surpassed the 0.50 threshold (Table 4), confirming convergent validity and internal consistency for reflective specification [44,45].

A 7-point Likert scale was used throughout the questionnaire, ranging from 1 = strongly disagree to 7 = strongly agree. Items were selected and, where necessary, adapted to ensure conceptual coherence, brevity, and clarity for respondents while maintaining validity.

Organisational Environmental Policy (EOP). This construct was measured using three adapted items from a broader 13-item scale developed by Ramus and Steger [46]. The selected items were chosen for their capacity to reflect tangible and objective organisational commitments, such as the implementation of environmental management systems and green purchasing practices, for instance. This scale has been previously validated in sustainability research [8,22,42].

Supervisory Commitment to the Environment (SCE). Three items were adapted from the original six-item scale by Erdogan et al. [33], which captures employees’ perceptions of their supervisors’ environmental awareness and pro-environmental behaviours. These items were selected to retain the coverage of both attitudinal and behavioural dimensions while reducing redundancy. This construct has been used and validated in recent research [47].

Coworkers’ Green Work Climate (CGWC). Based on Norton et al. [22], this construct was measured using three adapted items that capture perceptions of co-workers’ collective attention to and engagement with environmental practices. This scale highlights that climate refers specifically to observable practices and behavioural patterns within workgroups, as opposed to deeper cultural values [48].

Customers’ Environmental Attitudes (CEAs). Employees’ perceptions of customers’ environmental attitudes were assessed using three items developed and adapted on the basis of Norton et al. [22]. Items capture the extent to which customers are perceived as valuing, responding to, and acting in alignment with environmental initiatives during their stay. These perceptions represent an important normative pressure in service settings, particularly in sustainability-sensitive destinations [22].

Green Behavioural Intention (GBI). Three items were adopted from the validated scale by Norton et al. [26], grounded in the TPB. The items measure the intention to act sustainably in the workplace, covering both general environmental responsibility and the willingness to adopt specific sustainable practices. This construct has been widely applied in tourism and sustainability research [49,50].

Employee Green Behaviour—In-role (EGB-IR). This construct was measured using a three-item subscale from Bissing-Olson et al. [51], which captures environmentally responsible behaviours performed as part of one’s job responsibilities. These in-role behaviours are considered essential components of organisational sustainability performance, as they reflect the extent to which employees fulfil their tasks in line with environmental values [51]. The scale has been widely used in hospitality and organisational behaviour literature [5,38,52].

3.3. Sample

This study was conducted in the Canary Islands, one of Europe’s most visited mass tourism destinations, known for its environmental vulnerabilities and growing sustainability efforts. The archipelago hosts over 17 million tourists annually, offering a highly relevant context for analysing sustainable behaviour within the hospitality workforce. The Canary Islands is one the leading European NUTS 2 regions in terms of tourism establishments [53], and its mature hospitality industry, extensive tourism infrastructure, and high volume of international arrivals make it a particularly insightful context to examine the dynamics of employee green behaviour in hotels. Moreover, the Canary Islands share key characteristics with other coastal tourism regions globally—such as overdependence on tourism or resource-intensive operations—thus offering valuable implications for sustainability strategies in similar settings.

The sampling strategy followed a two-stage approach: hotels were the primary sampling unit (recruited through voluntary participation), and employees constituted the secondary unit (randomly invited to complete the questionnaire within each hotel). A total of 70 hotels participated in the survey. The sample included properties across three categories—1 to 3 stars, 4 stars, and 5 stars—and aimed to capture a representative picture of the regional hospitality sector. A comparison with the official hotel universe (

Table 2. Universe and primary sampling comparison.

Hotel Category	Number of Hotels
	Universe	Sample
1 to 3 stars	242	11
4 stars	293	39
5 stars	68	20
Total	603	70

) indicates that the distribution of establishments is broadly aligned with the sector’s structure. A total of 70 hotels participated in the survey. The sample included properties across three categories—1 to 3 stars, 4 stars, and 5 stars—and aimed to capture a representative picture of the regional hospitality sector. A comparison with the official hotel universe (Table 2) indicates that the distribution of establishments is broadly aligned with the sector’s structure.

Table 3. Secondary sampling structure.

Gender/Sex
Female	62.0%
Male	38.0%
Age
18–27 years	11.1%
28–37 years	24.7%
38–45 years	21.7%
>45 years	42.5%
Educational level
Basic education	24.5%
Upper secondary education	24.7%
Vocational education	22.9%
Higher education	27.8%
Seniority in the company
<3 years	45.5%
4–10 years	24.1%
11–20 years	16.5%
>20 years	13.9%
Hotel category
1 to 3 stars	8.9%
4 stars	55.1%
5 stars	36.0%
Department
Management, administration	9.7%
Food and Beverage (Bar, Restaurant, and Kitchen)	33.0%
Housekeeping	27.6%
Reception	14.3%
Technical service, maintenance	4.6%
Others	10.9%
Valid responses	497

presents the socio-demographic structure of the final sample. The data reveal that 46.4% of employees are between 28 and 45 years old, while 42.5% are over 45, reflecting the presence of a relatively mature workforce. Regarding educational attainment, 72.2% of respondents do not hold a higher education degree, which aligns with prevailing educational trends in the hospitality sector. The sample is also highly feminised, with women representing 62.0% of participants. This gender distribution is particularly prominent in core operational departments such as food and beverage (33.0%) and housekeeping (27.6%), which together account for over 60% of the workforce.

Initially, 530 responses were collected. After applying a three-level quality control process—which included checks for straightlining behaviour, the response time, and control questions—497 valid responses were retained for analysis. This final sample reflects a high level of data reliability, with no missing values.

To enhance generalisability, the sample was designed to include employees from diverse departments, including housekeeping, food and beverage, reception, technical services, and administration. This approach ensured that different functional roles and work conditions were represented in the dataset, acknowledging the heterogeneous nature of hotel labour structures.

A power analysis conducted using G*Power 3.1.9.4 [54] determined that a minimum sample size of approximately 300 participants would be required to detect medium-sized effects (f² = 0.15) in a structural equation model with up to five predictors at a 95% confidence level (α = 0.05) and power of 0.80. With a final sample of 497 valid responses, this study exceeds the recommended threshold, ensuring sufficient statistical power to reliably estimate path coefficients and test mediation effects.

To ensure the representativeness of the data and assess potential non-response bias, we calculated the overall survey response rate. Out of 1393 hotel employees initially contacted, 530 completed the questionnaire. After applying a rigorous data quality control process, 497 valid responses were retained for analysis, yielding an effective response rate of 35.7%. This figure aligns with typical response rates in organisational research, particularly in the hospitality sector, where rates between 30% and 40% are commonly reported [55].

To strengthen the methodological rigour of the survey, several strategies commonly recommended in organisational research were implemented. First, the questionnaire was distributed randomly among hotel employees, helping to prevent selection bias by ensuring that all eligible respondents had an equal opportunity to participate [56]. The survey remained open for a six-month period, which enhanced accessibility and reduced the risk of excluding employees with irregular schedules or high workloads [57]. Participation was entirely voluntary, and no incentives or managerial endorsements were used, minimising potential pressure or coercion and reducing the likelihood of social desirability bias in responses [58]. Furthermore, the dataset contains no missing values, as incomplete responses were filtered out during the data validation phase. This strengthens the internal reliability and ensures that selective non-response does not compromise the quality or completeness of the data.

Furthermore, this research complies with the ethical standards. Participation in the study was entirely voluntary, and informed consent was obtained electronically. Before beginning the questionnaire, participants received detailed information on the study’s purpose, anonymity assurances, data use, and their right to decline. No exclusion criteria were applied beyond data quality controls.

3.4. Data Analysis

The data were analysed using SmartPLS 4.1.0.2 [59], employing the Partial Least Squares (PLS) approach given the reflective nature of all constructs [60]. The path weighting scheme was applied, with standardised data, a stop criterion of 10⁻⁷, and a maximum of 300 iterations.

PLS-SEM was selected over CB-SEM due to its predictive focus, robustness to non-normality assumptions, and ability to handle complex models with multiple mediating relationships simultaneously [44]. This approach aligns with our research objectives of predicting employees’ green workplace behaviour and understanding the underlying mechanisms.

The measurement model evaluation included composite reliability, average variance extracted (AVE), and discriminant validity via the HTMT ratio with bootstrapped confidence intervals [61]. The structural model was assessed using path coefficients, R² values, and effect sizes (f²). Predictive relevance was analysed using the PLSpredict procedure with 10-fold cross-validation [62].

To control for potential common method bias (CMB), we applied several procedures. Harman’s single-factor test revealed that the first factor explained less than 50% of the total variance. Multicollinearity checks confirmed that all VIF values were below 3.0, and the inclusion of a latent marker variable showed no significant influence on the main constructs. These results indicate that CMB is not a critical concern in this study.

To assess the model’s out-of-sample predictive power, a PLSpredict procedure was performed. The results indicated that the PLS-SEM model outperformed a linear regression benchmark (LM) in terms of lower RMSE and MAE values for all key endogenous constructs, confirming its predictive relevance [62].

The univariate normality of all indicators was assessed through a skewness and kurtosis analysis. Results revealed that all indicators met the established criteria for PLS-SEM, with skewness values within the acceptable range of −2 to +2 (observed range: −1.38 to −0.22) and kurtosis values within the range of −7 to +7 (observed range: −0.66 to 3.36). Given that PLS-SEM is robust to moderate deviations from univariate normality, these results confirm the suitability of the data for the proposed analysis.

4. Results

4.1. Descriptive Analysis

Table 4. Descriptive analysis and results of the evaluation of the measurement model.

Reflective Mode A constructs		Mean	Standard Deviation	Loading	CR	AVE
GBI	Green Behavioural Intention				0.863	0.677
GBI1	I intend to act in an environmentally responsible manner in the workplace	5.84	1.124	0.841
GBI2	At work, I intend to engage in environmentally friendly behaviours	5.97	1.185	0.826
GBI3	I plan to carry out sustainable practices while I am at work	5.74	1.582	0.801
EOP	Environmental Organisational Policy				0.906	0.764
EOP1	My company has an environmental strategy and periodically publishes an environmental report	4.77	1.781	0.850
EOP2	My company uses an environmental management system	5.14	1.757	0.856
EOP3	My company takes environmental aspects into account when making decisions	5.43	1.817	0.914
SCE	Supervisor Commitment to the Environment				0.919	0.790
SCE1	My supervisor values and cares about environmental issues	4.94	1.801	0.857
SCE2	My supervisor is aware of the need to take action to protect the environment	5.20	1.766	0.864
SCE3	My supervisor considers the environmental impact of their actions and is committed to preserving and protecting the environment	5.70	1.939	0.943
CGWC	Coworkers’ Green Work Climate				0.931	0.818
CGWC1	My coworkers pay attention to existing environmental issues	4.38	1.686	0.875
CGWC2	My coworkers care about acting sustainably towards the environment and try to minimise environmental harm	4.49	1.724	0.890
CGWC3	My coworkers care about natural resources	4.99	1.800	0.948
CEA	Customers’ Environmental Attitude				0.911	0.772
CEA1	Customers staying at our hotel value the environmental actions carried out	4.35	1.763	0.873
CEA2	Customers at our hotel are environmentally sensitive	4.27	1.753	0.882
CEA3	Customers at our hotel behave sustainably during their stay	4.26	1.761	0.881
EGBIR	Employee Green Behaviour In Role				0.920	0.793
EGBIR1	I usually perform my work tasks in an environmentally responsible way	5.64	1.160	0.856
EGBIR2	I always fulfil the responsibilities assigned to my job position in an environmentally responsible manner	5.79	1.272	0.889
EGBIR3	I generally complete the tasks expected in an environmentally responsible way	6.11	1.293	0.925

Notes: CR: composite reliability (rho_c). AVE: average variance extracted.

presents the means and standard deviations of the measured constructs in the proposed model. The results offer preliminary insights into the levels of green behavioural intention, perceived sustainability context, and green behaviours reported by hotel employees.

The items measuring green behavioural intention (GBI) showed consistently high scores, reflecting a strong internal motivation to engage in environmentally responsible behaviours at work. The highest agreement was observed for the intention to perform sustainable actions in the workplace (GBI2: M = 5.97, SD = 1.19), followed closely by the intention to act responsibly (GBI1: M = 5.84, SD = 1.12) and to implement sustainable practices more broadly (GBI3: M = 5.74, SD = 1.58).

Regarding the corporate environmental policy (EOP), responses suggest moderate agreement with the environmental practices promoted by their hotels. The presence of an environmental strategy (EOP1: M = 4.77, SD = 1.78) and the use of an environmental management system (EOP2: M = 5.14, SD = 1.76) were acknowledged, although the scores suggest variations in the organisational commitment or visibility of these initiatives.

Employees’ perceptions of clients’ attitudes (CLIs) also yielded moderately high values. The belief that guests value environmental practices (CLI1: M = 5.77, SD = 1.32), show environmental sensitivity (CLI2: M = 5.63, SD = 1.39), and behave sustainably during their stay (CLI3: M = 5.30, SD = 1.43) indicates the presence of pro-sustainability social norms among clients.

In terms of co-workers’ green work climate (CGWC), employees reported that their colleagues demonstrate environmentally friendly behaviours and attitudes. Scores ranged from 5.36 to 5.67, with particularly high agreement based on the collective attention to environmental issues (CGWC1: M = 5.67, SD = 1.28) and concern about minimising environmental impact (CGWC2: M = 5.45, SD = 1.39).

Finally, perceptions of supervisors’ commitment to the environment (SCE) were also positive. Supervisors were described as environmentally aware (SCE2: M = 5.58, SD = 1.44), caring (SCE1: M = 5.46, SD = 1.42), and committed to protecting the environment in their actions (SCE3: M = 5.36, SD = 1.54), suggesting that leadership plays a supportive role in shaping sustainable behaviours.

As a reference, values between 5 and 7 on the 7-point Likert scale may be interpreted as moderate to high agreement, while values below 4 indicate a lack of agreement or perceived support. The overall means observed suggest that employees hold strong green behavioural intentions and perceive moderate to high levels of organisational and social support for sustainability. Nevertheless, the variability in standard deviations, particularly for constructs such as EOP and SCE, indicates the heterogeneity in respondents’ experiences, likely influenced by departmental differences or inconsistencies in managerial practices across hotels.

4.2. Evaluation of the Measurement Model

All constructs in the model were conceptualised as reflective and assessed according to the established criteria for Partial Least Squares Structural Equation Modelling (PLS-SEM) [44,63].

Table 5. Discriminant Validity (HTMT).

	CGWC	CEA	EGBIR	EOP	GBI	SCE
CGWC
CEA	0.454
EGBIR	0.544	0.420
EOP	0.649	0.489	0.590
GBI	0.386	0.294	0.634	0.464
SCE	0.732	0.473	0.585	0.843	0.452

reports the item loadings, composite reliability (CR), and average variance extracted (AVE) for each construct.

All item loadings surpassed the recommended threshold of 0.707 [64], supporting indicator reliability. The composite reliability values of the constructs ranged from 0.860 to 0.955, exceeding the 0.70 cut-off [60], and confirming strong internal consistency. In addition, all constructs met the 0.50 criterion for convergent validity, with AVE values ranging from 0.675 to 0.832 [45].

Discriminant validity was assessed using the Heterotrait–Monotrait ratio (HTMT). All HTMT values were below the conservative threshold of 0.85, demonstrating adequate discriminant validity between all construct pairs [61]. Specifically, the highest HTMT value observed was between CGWC and SCE (0.743), indicating that even closely related constructs remain empirically distinct.

In addition, multicollinearity was assessed through inner VIF values, which remained well below the critical value of 3.0, suggesting no collinearity concerns among the predictor variables [63]. Since all constructs were estimated reflectively, redundancy analysis was not required.

To further assess the discriminant validity, the Fornell–Larcker criterion was examined. For each construct, the square root of the AVE exceeded its correlations with other constructs, confirming adequate discriminant validity.

Finally, the model’s overall fit, evaluated through the Standardised Root Mean Square Residual (SRMR), was 0.044—below the 0.08 benchmark proposed for PLS path models [61]—indicating an acceptable global model fit.

These findings confirm the reliability, convergent validity, and discriminant validity of the measurement model, thereby supporting the robustness of the instrument for subsequent structural analysis.

4.3. Evaluation of the Structural Model

Figure 2 represents the structural model with standardised path coefficients and R² values for all endogenous constructs, illustrating the hypothesised relationships among the constructs and their indicators. To test the proposed paths, we applied bootstrapping with 5,000 subsamples, evaluating path coefficients, significance levels, bias-corrected confidence intervals, and effect sizes [44,63].

Table 6 shows the results of the structural model, testing all proposed hypotheses (H1–H13). The findings provide robust support for the theoretical model, confirming both the direct and indirect relationships among constructs.

H1 received strong support, as green behavioural intention (GBI) showed a significant direct effect on employees’ in-role green behaviour (EGB-IR) (β = 0.336, p < 0.001, 95% CI [0.232, 0.433]), confirming that individual motivation remains a powerful predictor of pro-environmental workplace actions. This relationship demonstrated a large effect size (f² = 0.166), indicating substantial practical significance.

H2 to H5 tested the influence of GBI on four contextual perceptions. All hypotheses were supported with large effect sizes: GBI positively predicted supervisors’ commitment to the environment (H2: β = 0.369, p < 0.001, f² = 0.157), environmental organisational policy (H5: β = 0.376, p < 0.001, f² = 0.164), coworkers’ green work climate (H3: β = 0.321, p < 0.001, f² = 0.115), and customers’ environmental attitudes (H4: β = 0.237, p < 0.001, f² = 0.059). These results confirm that employees with stronger green intentions tend to perceive greater alignment and support from their social and organisational environment.

H6, H8, H10, and H12 assessed the direct effects of contextual variables on green behaviour. All were statistically significant: supervisors’ commitment (H6: β = 0.131, p < 0.01, f² = 0.012), coworkers’ climate (H8: β = 0.158, p < 0.01, f² = 0.024), organisational policy (H12: β = 0.159, p < 0.001, f² = 0.020), and customers’ attitudes (H10: β = 0.099, p < 0.05, f² = 0.014). Although these relationships showed small to moderate effect sizes, their significance underscores their complementary role in enabling pro-environmental workplace conduct.

H7, H9, H11, and H13 tested the mediating effects of contextual factors in the GBI–EGB-IR relationship. All four mediation paths were significant with bias-corrected 95% confidence intervals excluding zero, confirming partial mediation. The Variance Accounted For (VAF) values indicate weak to moderate mediation effects: supervisors’ commitment (VAF = 12.5%), organisational policy (VAF = 15.2%), customers’ attitudes (VAF = 6.4%), and coworkers’ climate (VAF = 13.2%). These VAF values are consistent with other workplace behaviour studies, suggesting that contextual perceptions amplify the effect of intention without fully replacing its central role, validating the integrative approach of combining TPB and VBN perspectives.

While the structural results broadly confirm the expected relationships and align with prior research on pro-environmental workplace behaviours [6,12], some nuances deserve further reflection. For instance, the relatively small effect size of customer attitudes on employee green behaviour (β = 0.099, f² = 0.014) contrasts with studies that highlight clients as strong normative referents in hospitality contexts [65,66]. This may suggest that employees are more responsive to internal cues (e.g., coworker or supervisor behaviours) than to customer expectations, particularly when these are perceived as diffuse or inconsistent. Similarly, although all contextual variables partially mediated the GBI–EGB-IR relationship, the VAF values remained moderate, implying that while the social-organisational environment reinforces intention, it does not fully account for behaviour. These findings align with the TPB’s emphasis on intention as a proximal determinant but also resonate with VBN-derived insights on the limits of motivation in the absence of enabling conditions [13]. Future research could explore whether the strength of these mediations varies across hotel segments, job roles, or levels of environmental training received.

The predictive relevance (Q²) was assessed using the blindfolding procedure (omission distance = 7). All Q² values were positive, indicating acceptable predictive relevance for the endogenous constructs: EGB-IR (0.259), EOP (0.135), SCE (0.130), CGWC (0.099), and CLI (0.051).

Table 6. Results and hypothesis testing.

Hypothesis	Relationships	Path Coefficient	Sig.	Confidence Intervals	Confidence Intervals Bias	Supported	f-Square
H8	CGWC -> EGBIR	0.158	***	[0.051; 0.269]	[0.055; 0.273]	Yes/Yes	0.024
H10	CEA -> EGBIR	0.099	**	[0.015; 0.179]	[0.018; 0.181]	Yes/Yes	0.014
H12	EOP -> EGBIR	0.159	***	[0.056; 0.261]	[0.059; 0.265]	Yes/Yes	0.020
H3	GBI -> CGWC	0.321	***	[0.236; 0.402]	[0.234; 0.4]	Yes/Yes	0.115
H4	GBI -> CEA	0.237	***	[0.154; 0.318]	[0.15; 0.315]	Yes/Yes	0.059
H1	GBI -> EGBIR	0.336	***	[0.239; 0.439]	[0.232; 0.433]	Yes/Yes	0.166
H5	GBI -> EOP	0.376	***	[0.299; 0.453]	[0.293; 0.448]	Yes/Yes	0.164
H2	GBI -> SCE	0.369	***	[0.288; 0.45]	[0.284; 0.446]	Yes/Yes	0.157
H6	SCE -> EGBIR	0.131	**	[0.022; 0.244]	[0.019; 0.242]	Yes/Yes	0.012
Mediation hypothesis		Indirect effect	Sig.				VAF
H7	GBI -> SCE -> EGBIR	0.048	**	[0.008; 0.095]	[0.008; 0.095]	Yes/Yes	12.5%
H13	GBI -> EOP -> EGBIR	0.060	***	[0.021; 0.101]	[0.023; 0.103]	Yes/Yes	15.2%
H11	GBI -> CEA -> EGBIR	0.023	**	[0.004; 0.043]	[0.006; 0.046]	Yes/Yes	6.4%
H9	GBI -> CGWC -> EGBIR	0.051	**	[0.016; 0.088]	[0.019; 0.092]	Yes/Yes	13.2%
Model performance and predictive relevance	R-square	Q2predict
EGBIR	0.434	0.259
CGWC	0.103	0.099
CEA	0.056	0.051
EOP	0.141	0.135
SCE	0.136	0.130

Notes: bootstrapping, n = 5000 subsamples, significance * p < 0.05; ** p < 0.01; *** p < 0.001; ns: non-significant (one-tailed t Student). VAF: variance accounted for. Effect sizes f²: <0.005 low; <0.01 moderate; ≥0.025 high [67].

Table 6. Results and hypothesis testing.

Hypothesis	Relationships	Path Coefficient	Sig.	Confidence Intervals	Confidence Intervals Bias	Supported	f-Square
H8	CGWC -> EGBIR	0.158	***	[0.051; 0.269]	[0.055; 0.273]	Yes/Yes	0.024
H10	CEA -> EGBIR	0.099	**	[0.015; 0.179]	[0.018; 0.181]	Yes/Yes	0.014
H12	EOP -> EGBIR	0.159	***	[0.056; 0.261]	[0.059; 0.265]	Yes/Yes	0.020
H3	GBI -> CGWC	0.321	***	[0.236; 0.402]	[0.234; 0.4]	Yes/Yes	0.115
H4	GBI -> CEA	0.237	***	[0.154; 0.318]	[0.15; 0.315]	Yes/Yes	0.059
H1	GBI -> EGBIR	0.336	***	[0.239; 0.439]	[0.232; 0.433]	Yes/Yes	0.166
H5	GBI -> EOP	0.376	***	[0.299; 0.453]	[0.293; 0.448]	Yes/Yes	0.164
H2	GBI -> SCE	0.369	***	[0.288; 0.45]	[0.284; 0.446]	Yes/Yes	0.157
H6	SCE -> EGBIR	0.131	**	[0.022; 0.244]	[0.019; 0.242]	Yes/Yes	0.012
Mediation hypothesis		Indirect effect	Sig.				VAF
H7	GBI -> SCE -> EGBIR	0.048	**	[0.008; 0.095]	[0.008; 0.095]	Yes/Yes	12.5%
H13	GBI -> EOP -> EGBIR	0.060	***	[0.021; 0.101]	[0.023; 0.103]	Yes/Yes	15.2%
H11	GBI -> CEA -> EGBIR	0.023	**	[0.004; 0.043]	[0.006; 0.046]	Yes/Yes	6.4%
H9	GBI -> CGWC -> EGBIR	0.051	**	[0.016; 0.088]	[0.019; 0.092]	Yes/Yes	13.2%
Model performance and predictive relevance	R-square	Q2predict
EGBIR	0.434	0.259
CGWC	0.103	0.099
CEA	0.056	0.051
EOP	0.141	0.135
SCE	0.136	0.130

Notes: bootstrapping, n = 5000 subsamples, significance * p < 0.05; ** p < 0.01; *** p < 0.001; ns: non-significant (one-tailed t Student). VAF: variance accounted for. Effect sizes f²: <0.005 low; <0.01 moderate; ≥0.025 high [67].

The structural model explains 43.4% of the variance in EGB-IR, with the additional explained variance for mediating constructs ranging from 5.6% (customers’ attitudes) to 14.1% (organisational policy). Predictive relevance was assessed using the PLSpredict procedure [62], which yielded Q² values above zero for all endogenous constructs, confirming acceptable predictive capability: EGB-IR (Q² = 0.259), coworkers’ climate (Q² = 0.099), organisational policy (Q² = 0.135), supervisors’ commitment (Q² = 0.130), and customers’ attitudes (Q² = 0.051). When comparing the PLS-SEM model against the linear regression model (LM) benchmark, 8 of 15 indicators showed lower RMSE (Root Mean Square Error) values and 10 of 15 indicators presented lower MAE (Mean Absolute Error) values, indicating that the majority of indicators outperform the linear benchmark. The EGB-IR and supervisors’ commitment constructs demonstrated the strongest predictive capability with pronounced differences favouring PLS-SEM (MAE differences up to −0.068), while the coworkers’ climate construct presented more modest but still competitive performance. Overall, the model demonstrates moderate to high predictive power according to Shmueli et al.’s [62] criteria, confirming that the PLS-SEM approach adds predictive value compared to simpler statistical methods and is appropriate for predictive purposes.

5. Discussion and Implications

By bridging TPB and VBN theories and testing them empirically in the hospitality sector, this research offers a more comprehensive understanding of how employees’ sustainable behaviours emerge. It demonstrates that promoting green intentions is a necessary but insufficient condition: creating supportive social environments and robust organisational structures is crucial for translating pro-environmental motivations into consistent workplace actions.

This study provides empirical support for the key role of employees’ green behavioural intention (GBI) in fostering sustainable behaviours in the hospitality sector. Consistent with the TPB [9], the results confirm that GBI is a strong and significant predictor of in-role green behaviour (EGB-IR). This finding aligns with previous research highlighting intention as the primary driver of environmentally responsible actions in service settings [6,10,15]. However, this study further contributes to our understanding by demonstrating that the translation of intention into behaviour is not isolated from the social and organisational context.

By integrating the VBN theory [13], this research acknowledges that green workplace behaviour is not just a rational, planned outcome but also influenced by internalised values, perceived norms, and moral obligations activated through external stimuli. The significant relationships observed between GBI and contextual perceptions—supervisors’ commitment to the environment (SCE), coworkers’ green work climate (CGWC), customers’ environmental attitudes (CEA), and perceived environmental organisational policy (EOP)—suggest that employees with stronger sustainability intentions are also more sensitive to detecting and valuing supportive environmental cues in their work environment. These findings align with prior literature arguing for enriched models of sustainable behaviour that incorporate both personal and contextual influences [2,12].

The results confirm H1, as employees’ green behavioural intention (GBI) significantly predicts their in-role green behaviour (EGB-IR). This finding aligns with previous studies [6,10,15] and reinforces the centrality of intention as a key driver of workplace sustainability conduct.

The results also support H2 to H5, demonstrating that GBI positively influences employees’ perceptions of supervisors’ commitment (H2), coworkers’ green work climate (H3), customers’ environmental attitudes (H4), and environmental organisational policy (H5). This suggests that employees with stronger green intentions are also more attuned to sustainability signals in their work environments, supporting the relevance of combining TPB with VBN to account for contextual interpretation mechanisms.

With regard to the mediation pathways, H7 to H10 were confirmed. Supervisors, coworkers, customers, and organisational policy all partially mediate the effect of GBI on EGB-IR, supporting the hypothesis that contextual variables do not substitute but rather reinforce the impact of personal intention. These findings expand upon prior critiques of TPB’s limited scope by empirically validating the need for integrated behavioural models [16,19].

The direct effects of contextual mediators on green behaviour (H6, H8, H10, and H12) were also significant, reinforcing the importance of vertical, horizontal, external, and structural factors in shaping employees’ sustainable actions. Notably, H13 is supported as well, showing that formal environmental policies enhance the realisation of individual pro-environmental intentions through increased perceived behavioural control and value alignment.

The partial mediation pattern observed—with all mediators having a significant but not exclusive role—suggests that sustainable workplace behaviour emerges from a complex interaction between personal motives and social-organisational enablers. This supports the development of more holistic models that reflect the layered nature of sustainability engagement in service settings.

Furthermore, the effect sizes and explained variance observed highlight the practical relevance of these dynamics. GBI exhibited the strongest direct effect on green workplace behaviour, but the contributions of organisational policies, supervisors’ leadership, coworker climate, and customer attitudes—though smaller—were statistically and practically meaningful. The model explains 46.4% of the variance in employees’ green behaviour, a considerable proportion compared to prior research in organisational sustainability contexts [3,8].

In sum, all thirteen hypotheses (H1–H13) were empirically supported, confirming both the direct and indirect relationships among intention, contextual factors, and green behaviour. This enhances the theoretical robustness of the proposed model and affirms the value of a dual-theory approach to understanding pro-environmental action in hospitality. Encouraging individual environmental responsibility remains critical, but it must be accompanied by visible leadership commitment, the importance of nurturing a green work climate among peers, responsiveness to customers’ environmental expectations, and the implementation of formal environmental policies. These strategies not only enhance behavioural alignment but also contribute to creating an organisational culture where sustainable practices become the normative standard [22,35].

6. Limitations and Future Lines of Research

Although the current study did not include a multi-group analysis (MGA) or control variables, future research should consider subgroup comparisons (e.g., by gender, age, or department) to explore whether the observed relationships hold across different employee profiles. The rich demographic dataset offers opportunities for the further exploration of moderating effects.

Moreover, the study relied exclusively on employee self-reported data, which may introduce perceptual biases, particularly for constructs such as customers’ environmental attitudes or organisational environmental commitment. Although the survey ensured anonymity and used validated scales to mitigate social desirability bias, future research would benefit from methodological triangulation. For instance, collecting complementary data from customers, incorporating managerial evaluations, or using third-party sustainability audits could help validate employee perceptions and strengthen the methodological soundness of similar studies. Combining multiple perspectives would also offer a more comprehensive view of the socio-organisational mechanisms influencing green workplace behaviour.

Nevertheless, although the model offers robust explanatory and predictive power, the cross-sectional nature of the data restricts causal interpretations. Longitudinal research would provide deeper insights into the stability and evolution of green behaviours over time. Further studies are also encouraged to explore additional moderators or mediators that might influence the translation of green intention into behaviour. Variables such as environmental identity [22], green transformational leadership [68], or perceptions of organisational justice could offer valuable extensions. Moreover, examining how employees’ sustainability behaviours evolve in response to corporate environmental initiatives over time could provide a more dynamic understanding of sustainable change processes [21].

Finally, although the study offers valuable insights for the hospitality sector, its geographic and cultural context must be acknowledged. The research was conducted in the Canary Islands, a mature mass tourism destination with strong sustainability pressures and institutional commitments to environmental protection. The Canary Islands is one the leading European NUTS 2 regions in terms of tourism establishments [53], sharing characteristics and challenges with most developed coastal tourism regions in Europe and beyond. Therefore, the results can provide very useful insights for many regions around the world. However, the transferability of results to some regions, particularly those with different tourism models (e.g., rural, urban, cultural, nature based, etc.), labour market dynamics, or cultural values, may be limited as the differences may influence how employees interpret and act upon sustainability cues. Future research could explore the proposed model in alternative contexts—such as emerging tourism markets, urban hotel settings, or non-European regions—to examine potential variations. Comparative cross-cultural or cross-sectoral designs would help to determine whether the observed mediation patterns are consistent or contingent on contextual conditions.

7. Conclusions

This study underscores that achieving meaningful progress towards sustainability in the hospitality sector demands a systemic approach, where employees’ pro-environmental intentions are actively supported, validated, and reinforced by their social and organisational environment. By bridging TPB and VBN perspectives, the research provides a more holistic framework for understanding and promoting sustainable workplace behaviours.

The results advance the understanding of sustainable workplace behaviours by proposing and empirically validating an extended model that integrates the TPB and the VBN theory [9,13] in the hospitality sector. The findings reaffirm the central role of employees’ green behavioural intention (GBI) as a key driver of in-role green behaviour (EGB-IR), while also highlighting the crucial influence of contextual factors—supervisors’ commitment, coworkers’ green work climate, clients’ environmental attitudes, and environmental organisational policies—in reinforcing or facilitating this relationship [6,10,19].

By demonstrating the partial mediation effects of these contextual variables, the study responds to recent calls for enriched models that move beyond purely individualistic explanations of sustainable behaviour, recognising the embeddedness of green actions within social and organisational ecosystems [2,12]. It supports the idea that fostering sustainability in service industries requires the creation of multi-level supportive structures, rather than relying only on individual motivation [3,21].

From a practical standpoint, the findings offer concrete guidance for hospitality organisations seeking to foster green workplace behaviour. First, sustainability-related training programmes should go beyond technical knowledge and actively promote environmental values and personal moral engagement. Second, developing environmental leadership at all hierarchical levels—particularly among supervisors and middle managers—can enhance the perceived legitimacy and consistency of green expectations. Third, peer support mechanisms, such as green teams or recognition programmes, can nurture a shared climate of environmental responsibility. Finally, incentive systems aligned with sustainability goals—including non-monetary rewards, gamified challenges, or performance appraisals incorporating environmental indicators—may further motivate employees to act in accordance with their green intentions. Taken together, these strategies can help institutionalise sustainability into the organisational culture, reinforcing intention–behaviour consistency and facilitating long-term change [34].

Beyond the organisational sphere, the findings also carry relevant implications for policymakers and sustainability certification bodies. These actors play a key role in shaping industry standards and expectations. Our results suggest that institutional efforts should move beyond compliance-based approaches to actively foster workplace cultures that support green behaviour from the bottom up. For instance, certification schemes could incorporate psychosocial indicators (e.g., employee engagement with sustainability, perceived leadership support) into their evaluation criteria. Likewise, public policies promoting sustainable tourism should consider incentives not only for infrastructure or environmental reporting, but also for capacity building, training, and leadership development aimed at cultivating pro-environmental behaviours within hotel teams.