Driving green shipping through psychological capital and organizational climate to bridge the awareness-action gap for enhanced environmental policy outcomes

1 Introduction

Global climate change is one of the most important existential challenges of our time. Promoting green low carbon transition and attaining “carbon peaking and carbon neutrality” are goals of the international climate change conference (Wang et al., 2024; Xiao et al., 2024). In this context, green transformation is particularly important for the shipping industry, the lifeblood of the global economy. It accounts for about 2.89% of global greenhouse gas emissions, and under increasingly stringent environmental regulations from the International Maritime Organization (IMO) and greening global supply chains, its performance has gone from a “nice-to-have” for corporate image enhancement to a “must-have” for compliance, competitiveness, and license for sustainable operation (Wang, 2025).

The successful transformation of enterprises is not only dependent on technological innovations like the Energy Efficiency Design Index (EEDI), but more deeply on people (Stec et al., 2021). However, a common and persistent problem in organizational management practice is the value-action gap which severely limits the implementation of environmental strategies (Wang et al., 2023; Xiao et al., 2025). The gap is being widened in the shipping industry. Seafarers work for extended periods in isolated, high-pressure, and resource-constrained “floating territories”, and shore staff are the hub of a complex logistical support system. Whether in energy-saving operations at sea or resource management on shore, the pro-environmental behavior of these individuals constitutes the micro-foundation of a company’s environmental performance (Hoang et al., 2022; Xiao et al., 2022). Understanding and effectively bridging the awareness-action gap between shipboard and shore-based employees is crucial for the sustainable development of the shipping industry.

The consensus is that an individual’s pro-environmental values are the main driving force behind their environmental behavior (Gatersleben et al., 2014). However, translation of values (as a stable belief) into specific and sustained actions is not automatic, but a psychological process in uncertainty. Classic theories such as Theory of Planned Behavior (TPB) have shown that cognitive factors such as attitudes and subjective norms play a role in pro-environmental behavior, but they do not explain how much psychological resources individuals need to do pro-environmental behavior, or how they match up with external environment (Ajzen, 1991; Steinmetz et al., 2016). Particularly, there is no mechanistic study of whether the organization “activates” and “amplifies” or inhibits and “depletes” translation of psychological resources. This gap is the starting point and the main contribution of this study.

To explore the “black box” of translating awareness into action, we introduce two constructs: psychological capital, as defined by an individual’s positive psychological state, characterized by self-efficacy, optimism, hope and resilience developed through personal growth, serves as the internal motivational and capability resource that employees need to bridge the gap (Cole et al., 2009). It allows employees to face challenges confidently, pursue goals with hope and optimism, and recover from setbacks resiliently. In contrast, according to Social Cognitive Theory, behavior is the product of continuous interaction between the individual and the environment (Bandura, 1998; Schunk and DiBenedetto, 2020). Organizational green climate, characterized as employees’ shared perceptions of the extent to which their organization values, supports and rewards the protection of the environment, is the external situational context and signaling system shaping behavior (Zientara and Zamojska, 2018). A strong green climate provides needed policies, resources and incentives, but also, through its signaling effect, stimulates employees’ environmental efforts with meaning, and enables them to translate psychological capital to action.

In this study, we propose and test a moderated mediation model that is based on internal psychological processes and external boundary conditions of behavior generation that provides a more general explanation of the “awareness-action gap”: First, the direct impact of environmental values on the pro-environmental behavior of shipboard and shore employees; Second, the mediating role of psychological capital: whether environmental values promote the psychological capital of employees and promotes pro-environmental behavior; Third, whether organizational green climate positively moderates the promotion effect of environmental values on psychological capital, and whether it moderates the translation efficiency of psychological capital into pro-environmental behavior.

Combining value-action theory, the Theory of Planned Behavior, and the Conservation of Resources theory, we construct an “individual-situation” interaction theory (Ajzen, 1991). We also further understand the complex mechanism behind values-behavior translation and extend positive organizational behavior theory application to environmental management. Practically, the results will provide shipping industry managers a clear direction for action: achieving a green transition requires moving beyond traditional “preaching-style” environmental advocacy towards a systematic management project incorporating values guidance, psychological capital empowerment and organizational climate support. This will provide precise scientific evidence for human resource management and organizational intervention strategies.

The next chapter will summarize relevant work and identify gaps. Section 3 will present the theoretical model and research hypotheses. Section 4 will describe research methodology. Section 5 will present the data analysis results. Section 6 will discuss results and Section 7 will conclude the study and suggest directions for future work.

2 Literature review and theoretical foundations

2.1 Employee pro-environmental behavior and its organizational implications

To establish the theoretical basis of this work we first define the main dependent variable. Employee pro-environmental behavior (EPEB; hereafter PEB) refers to actions taken by employees in the workplace to avoid environmental harm or contribute to environmental sustainability (Yuriev et al., 2018). The term represents a shift from technological and institutional to micro-level behavior, recognizing employees as active agents rather than passive executors of organizational environmental performance.

Some early studies treated such behavior as a single whole, whereas recent studies emphasized its multidimensional nature. Among the most influential classifications is the task-related versus voluntary behavior dichotomy. Task-related pro-environmental behavior is explicitly stated in job descriptions and organizational rules, and its performance is an in-role requirement, such as seafarers complying with ballast water management rules (Fatima et al., 2023). Voluntary pro-environmental behavior is not a role requirement, but rather an action taken by employees, such as reporting leakage or advocating green procurement. While driving mechanisms may differ between these two types, the latter usually more reflect an organization’s “green DNA” and employees’ psychological identification, which plays an important role in shaping a sustainable organization culture.

It is worth noting that in recent years, it has been suggested that the boundary between task-related and voluntary behavior may be fluid rather than fixed in practice (Fatima et al., 2023). In high-reliability industries such as shipping, where safety and environment are very strict, many actions considered “voluntary” in a typical office environment may be included into standard operating procedures on ships as “task-related” behaviors. This change of behavior roles by industry norms suggests caution in scale design and interpretation of results.

Mature scales were adopted and adapted to ship-based and shore-based work scenarios in the shipping industry to ensure measurement validity. Clarifying the meaning of EPEB provides a target for studying its driving mechanism, which is often deeply embedded in a person’s stable value system.

2.2 Environmental values as the ethical foundation of pro-environmental behavior

After defining the subject, it is necessary to trace its drivers. Environmental values (EV) refer to beliefs about the value of nature and human relationship to nature and are the basis of individual behavior towards nature (Torkar and Bogner, 2019). They originate from Schwartz’s circular model, where “self-transcendence” values (universalism: concern for the environment and the welfare of others) are positively related to pro-environmental behavior, and “self-enhancement” values (power: emphasis on dominance over people and resources) are often negative.

Operationally, the New Ecological Paradigm (NEP) scale is the gold standard for measuring mainstream individual values (Dunlap, 2008). It represents a paradigm shift from anthropocentrism to ecocentrism. Although NEP scale is widely used, it measures primarily the belief in ecocentric principles (biospheric values) in an individual. Previous work, such as Stern et al.’s Value-Belief-Norm (VBN) theory distinguishes the different mechanisms through which “egoistic”, “altruistic” and “biospheric” values influence behavior (Wynveen et al., 2015). We focus on broad environmental values, whose meaning is more closely linked to “biospheric” values, the belief that nature has value as a whole, not as a tool for human service.

Numerous studies show that environmental values are a significant but weak predictor of pro-environmental behavior. This weak association indicates, therefore, that values have a “gray area” of mediators and moderators between values and behavior (Xiao et al., 2025). Values explain behavior, but do not fully answer the question of how it can be done or under what circumstances it will be done. These theories direct us to explore the key psychological bridge between values and behavior.

2.3 Psychological capital as a positive psychological resource for bridging the awareness-action gap

In order to understand the translation from “values to behavior” we introduce a mediator. Psychological capital (PC) is a measurable, developable, and performance-related positive psychological state manifested in the growth period of an individual. It includes self-efficacy (confidence to face environmental challenges), hope (pathways and willpower to achieve environmental goals), resilience (ability to cope with environmental challenges), and optimism (positive recognition of environmental results) (Eid et al., 2012).

The psychological capital introduced here provides a positive explanation of the “awareness-action gap”. According to the Conservation of Resources (COR) theory, individuals strive to obtain, retain and protect valuable resources (Hobfoll, 2001). Employees with strong environmental values have the motivation to do something (a kind of cognitive resource) first. To translate this motivation into action requires additional personal resources (e.g. time, effort) and may fail. At this point psychological capital as a reservoir of positive resources makes employees more willing and capable of making such investments: self-efficacy makes them believe they can succeed; hope helps them find ways; optimism allows them to anticipate positive results; and resilience makes them not give up easily when faced with resistance. This improves their self-efficacy, sustains their hope and optimism regarding green goal attainment, and strengthens their resilience to overcome setbacks, supported by inner convictions.

This mediating mechanism has been previously supported by Gkorezis, who found that employees’ environmental spirit positively influences their environmental behavior through psychological capital and Zhang et al. studies showing that psychological capital plays an important role between environmental concern and green behavior (Gkorezis, 2015). These studies provide direct empirical references for positioning psychological capital as the main mediator between environmental values and behavior in this study. Therefore, we propose that environmental values indirectly promote pro-environmental behavior by cultivating and enhancing employees’ psychological capital.

2.4 Organizational green climate as an empowering context

Individual psychological processes do not occur in isolation but are strongly affected by organizational context. Organizational green climate (OGC) refers to the perceptions of members of the organization about policies, procedures, practices, and expected behavior on environmental protection (Norton et al., 2014). It is a particular example of organizational climate in the environment and is the fundamental definition of the “environment” factor in Social Cognitive Theory (Bandura, 1991).

A strong green climate is a strong situational signal to employees, providing clear behavior norms, resource support (e.g. facilities, training) and consistent incentive feedback. It directs and regulates employee behavior, while also interacting with individual psychological resources. In addition, from a multilevel theory perspective, organizational green climate is a unit property and interacts with individual psychological capital jointly to influence individual behavior. This reflects the core tenet of Social Cognitive Theory that contextual factors and individual characteristics interact reciprocally to shape behavior (Bandura, 2001). In this view, a supportive organizational climate does not suppress individual differences but rather provides the enabling conditions for individual psychological resources such as psychological capital to be effectively expressed and translated into action.

An organizational green climate can empower psychological capital in two ways: First, as a contextual resource, it can support positive effects of environmental values on psychological capital (for example, when an organization strongly supports environmental protection, employees’ environmental beliefs become firmer, causing more hope and optimism). Second, it can help moderate predictor effects of psychological capital on behavior (even if employees are confident and hopeful, their behavior may not be feasible if the organization lacks necessary support; conversely, psychological capital can be released to its fullest extent in a supportive climate). Thus, organizational green climate is an important boundary condition for understanding when and how psychological capital can translate into behavior.

2.5 Literature review and theoretical framework development

Through systematic review and critical analysis of the literature, we have seen the development of these research areas and their shortcomings: First, we agree on measuring the multidimensional EPEB behavior of workers but its specific manifestations and measurements in special industries like shipping remain understudied. Second, environmental values have been known as a driving force of behavior, but not much research has been done to explain how they influence behavior, especially positive psychological states. Third, psychological capital, a new concept of positive psychology, has potential to explain workplace pro-environmental behavior, but no research has been published on how the interaction with organizational context is still in its early stages. Fourth, while we acknowledge the importance of organizational green climate, most studies focus on its direct effects or simple moderating roles. However, no research has been done to investigate how it interacts with individual psychological resources to jointly influence the pathway from cognition to behavior.

In summary, three main research gaps are identified in the present work. First, there is a lack of contextualized investigation, as no empirical study has systematically examined the driving mechanisms of employee pro-environmental behavior in industries characterized by high isolation and high reliability demands, such as shipping. Second, theoretical integration remains insufficient: most existing models either focus on isolated bivariate relationships or rely on single-mechanism frameworks, and few have simultaneously incorporated distal values, proximal psychological resources, and macro-level organizational contexts. Third, mechanistic depth is limited regarding the core question of how individual psychological resources interact with organizational context. Most studies have stopped at detecting interaction effects without further testing whether such interactions occur at the first stage of mediation (values → psychological capital), the second stage (psychological capital → behavior), or the entire mediating process, indicating that rigorous tests of moderated mediation models are still lacking.

In this sense, the conceptual framework and thesis presented in this study are summarized: We propose a moderated mediation model. The main idea is that employees’ environmental values promote pro-environmental behavior by increasing their psychological capital, and the organizational green climate moderates both stages of mediation path positively. The structural differences in work situations among ship and shore employees provide an industry context basis for further discussions on the applicability boundaries of theoretical models in different groups. In Section 3, we will elaborate some research hypotheses based on this framework. This conceptualization is not only a generalization and extension of existing research, but also a significant test of EPEB research in a critical industry.

3 Theoretical model and research hypotheses

3.1 Theoretical foundations and model development

This section aims to construct a framework for analyzing internal mechanisms and boundary conditions under which environmental values influence the pro-environmental behavior of shipboard and shore workers. The framework developed in this paper is based on the underlying problem of the awareness-action gap. Value-Act theory and the Theory of Planned Behavior provide a meta-theoretic framework for understanding the nature of the gap: stable environmental values must be converted into behavioral intentions and then into actual actions through more specific intermediate constructs like attitudes, subjective norms, perceived behavior control (Ajzen, 1991; La Barbera and Ajzen, 2020). The transformation process inherently has the risk of failure.

To explore this “transformation process” and to uncover its internal processes and external conditions, we combine Social Cognitive Theory and Conservation of Resources Theory which form the basis of “psychological capital-organizational climate” synergy model (Bandura, 2001; Hobfoll, 2001). Triadic reciprocal determinism of Social Cognitive Theory suggests that there is continuous, bidirectional interaction between environmental values and psychological capital, pro-environmental behavior and organizational green climate. Behavior is the result of the synergy between individual psychological resources and environment. Conservation of Resources Theory provides a motivational explanation of this synergy: individuals are motivated to acquire, retain and protect valuable resources (Hobfoll, 2001). Environmental values are cognitive resources, psychological capital is a positive psychological resource, and organizational green climate is a crucial contextual resource. Whether an employee is willing to put their cognitive resource into action using their psychological resource depends on whether they expect this investment will be protected and rewarded in a supportive environment, rather than reduce the resource.

Finally, we propose a moderated mediation model. The idea is that employees’ environmental values promote their pro-environmental behavior by enhancing their psychological capital. Meanwhile, the organizational environment supports both the cultivation effect of environmental values on psychological capital and the translation efficiency of psychological capital into pro-environmental behavior, thereby positively moderating this mediation path. In this model, the following research hypotheses are proposed (see Figure 1).

3.2 Research hypotheses

3.2.1 Direct effect hypotheses

H1: Environmental values have positive effect on employees’ pro-environmental behavior.

Value action theory states that a person’s values are the basic guide for his/her attitudes and behavior. Employees who hold strong pro-environmental values understand environmental protection as a moral and behavioral standard and engage more actively in environmental behavior at work, whether or not they are required to do so.

H2: Environmental values have a significant positive effect on employees’ psychological capital.

Conservation of Resources theory (COR) holds external information and challenges in accordance with one’s values can be used to gain valuable resources. When employees believe in environmental protection, environmental tasks at work are goal pursuits. This improves their self-efficacy, sustains their hope and optimism regarding green goal attainment, and strengthens their resilience to overcome setbacks, supported by inner convictions.

H3: Psychological capital positively affects employees’ pro-environmental behavior.

COR theory assumes that individuals with more resources are more willing to invest resources and more prepared to tackle challenges. Employees with more psychological capital have more psychological resources to manage the possible additional costs (e.g. time, effort) and uncertainties associated with pro-environmental behaviors. Self-efficacy allows them to act with confidence; hope allows them to find and implement different ways to achieve environmental goals; optimism allows them to expect positive behaviors regularly; resilience allows them to persist rather than easily give up when their actions are hindered or fail.

3.2.2 Mediation effect hypothesis

H4: Psychological capital mediates the relationship between environmental values and pro-environmental behavior.

Based on the logic chain of H2 and H3, we propose that environmental values influence pro-environmental behavior not only directly but also indirectly by shaping and improving a set of key positive psychological capacities (psychological capital). In other words, environmental values provide the “why to act,” while psychological capital supplies the motivation and strategies for “being able to do it and persevere.” Both are essential components of the psychological pathway to bridge the “awareness-action gap.”

3.2.3 Moderation effect hypotheses

H5: Organizational green climate positively moderates the relationship between environmental values and psychological capital.

The positive effect of environmental values on psychological capital is stronger when organizational green climate is high than when it is low. Based on the environment to person pathway in Social Cognitive Theory, a strongly supportive organizational green climate confirms and reinforces the correctness and importance of employees’ personal environmental values at the organizational level by providing clear policies, resources, signals and positive incentives (Bandura, 2001). When employees see a high alignment between their personal values and direction of the organization, their environmental convictions are firmer, their hope and optimism for environmental goals are stronger and their self-efficacy is increased by organization support. So psychological capital is activated, sustained and elevated by organization support.

H6: Organizational green climate positively moderates the relationship between psychological capital and pro-environmental behavior.

The positive effect of psychological capital on pro-environmental behavior is stronger when organizational green climate is high than when it is low. Based on person and environment interaction on behavior in Social Cognitive Theory, organizational green climate, as a key contextual resource, can create synergies with individual psychological resources (Bandura, 2001). In a strong green climate, organization provides “smooth channels” and “empowering platforms” (e.g. convenient environmental facilities, error tolerant culture encouraging innovation, timely recognition, and rewards for environmental contributions) for translating psychological capital into concrete actions. Consequently, the confidence, hope, optimism and resilience of employees with high psychological capital can be translated more smoothly, fully and efficiently into real actions in such a “highly supportive environment”.

3.2.4 Moderated mediation effect hypothesis

H7: Organizational green climate moderates the indirect effect of psychological capital (i.e. moderated mediation model).

This mediating effect (environmental values → psychological capital → pro-environmental behavior) is stronger when organizational green climate is high and weaker when it is low. This is the core hypothesis. Organizational green climate moderates both the first stage (environmental values→psychological capital, H5) and second stage (psychological capital→pro-environmental behavior, H6) of the mediation path, this means that the strength of the whole “environmental values→psychological capital→pro-environmental behavior” mediation path is not fixed, but depends strongly on the boundary condition of organizational green climate: only when employees have positive environmental psychological resources (high psychological capital) and their environment provides adequate support and incentives (high green climate) can they translate their environmental values more effectively and sustainably into pro-environmental behaviors. In this study, employee job type (seafarer vs. shore employee) was only included as a demographic variable in the control variables.

4 Research methodology

4.1 Research sample and data collection

4.1.1 Research subjects and sampling design

The objective of this study was to focus on practitioners in China’s shipping industry: seafarers and shore-based workers. Due to the particularities of the industry and the difficulty in collecting samples, a multi-channel convenience sampling strategy was adopted through the following channels: (1) collaboration with five shipping companies of different sizes across East China, North China, and South China, with questionnaires distributed to their employees with assistance from human resources departments; (2) online questionnaire distribution via platforms of the Chinese Institute of Navigation and shipping associations; and (3) the sample covered vessels operating on both international and domestic routes, various vessel types (container ships, bulk carriers, tankers, etc.), and diverse shore-based departments.

4.1.2 Data collection procedure and quality control

Data were collected between June and August 2025 using a combination of online and offline approaches. For the online survey, encrypted questionnaire links were generated via a professional survey platform and distributed to target respondents through the human resources departments of the participating companies. For the offline component, paper-based questionnaires were administered to seafarers during vessel port calls. The research protocol was approved by the relevant institutional ethics committee.

Several quality control measures were implemented to ensure data integrity. Digital fingerprinting was employed to identify and eliminate duplicate submissions. Two attention-check items were embedded in the questionnaire (e.g., “Please select the ‘Somewhat Agree’ option for this question”) to screen out inattentive respondents. A minimum completion time threshold was set, with responses completed in less than three minutes deemed invalid. Additionally, response patterns were examined to exclude systematic or careless answering styles, such as straightlining or repetitive patterning.

To mitigate potential contextual effects arising from the use of different survey modes, both versions were introduced with identical instructions and attention-check items, with anonymity and confidentiality emphasized to ensure psychological safety and response comparability.

4.1.3 Sample characteristics

We first distributed 600 questionnaires and received 542 responses. After data cleaning, 463 valid responses were found, and we have an effective response rate of 81.0%. Non-respondents showed no significant differences in age distribution or gender distribution compared to final sample. Details of the sample are shown in Table 1.

Table 1. Demographic Characteristics of the Sample (N = 463).

After cleaning the raw data by removing duplicate submissions, patterned responses, and questionnaires failing the attention checks, we finally obtained 463 valid responses for all subsequent analyses. A post-hoc statistical power analysis conducted using G*Power 3.1 indicated that, at a significance level of 0.05, the statistical power for detecting a medium effect size (f² = 0.15) was 0.98, well above the recommended threshold of 0.80, confirming that the sample size was adequate.

4.2 Measurement instruments

All scales underwent a rigorous cross-cultural adaptation process, including translation, back-translation, expert panel review, and pilot testing, to ensure their applicability in the Chinese shipping industry context. All items were rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

4.2.1 Environmental values

The new NEP scale by Dunlap et al. was adopted. Items analyses based on pilot test data (n = 65) show that critical ratios for all items are significant (p < 0.001) (Dunlap et al., 2000). Confirmatory Factor Analysis (CFA) shows that models fit well (χ²/df=2.86, CFI = 0.93, TLI = 0.91, RMSEA = 0.07). Factor loadings for the 15 items ranged from 0.58 to 0.79, Cronbach coefficient = 0.86.

4.2.2 Psychological capital

The simplified Chinese version of the Psychological Capital Questionnaire by Luthans et al. was used (Luthans et al., 2007). It is reliable and valid in China. The four dimensions were confirmed with CFA (χ²/df = 2.42, CFI = 0.95, TLI = 0.93, RMSEA = 0.06) Factor loadings for 12 items ranged from 0.63 to 0.82, Cronbach coefficient = 0.92, and coefficients for each dimension are 0.83 to 0.88.

4.2.3 Organizational green climate

The scale developed by Norton et al. was used (Norton et al., 2014). A panel of shipping industry experts (n = 5) assessed its content validity and obtained Scale-level Content Validity Index/Average (S-CVI/Ave) of 0.92. CFA supported a unidimensional structure with good fit (χ²/df = 3.15, CFI = 0.94, TLI = 0.92, RMSEA = 0.07). Factor loadings for the 6 items were 0.68-0.91, Cronbach’s coefficient 0.89.

4.2.4 Pro-environmental behavior

Graves et al. scaled to the nature of shipping (Greaves et al., 2013). Exploratory Factor Analysis (EFA) on a sub-sampled (n = 243) extracted two dimensions: Task-related behavior (6 items) and Voluntary environmental behavior (4 items) with explained variance of 61.3%. CFA on the same sub-sampled confirmed the two-factor structure (χ²/df = 2.78, CFI = 0.94, TLI = 0.92, RMSEA = 0.06). Cronbach’s coefficient was 0.90.

4.2.5 Control variables

Based on the literature review, five control variables that could influence pro-environmental behavior were selected: age, gender, education level, years of service, and position type (seafarer vs. shore employee). These variables were entered as covariates in subsequent analyses.

4.3 Data analysis strategy

In this study, quantitative analysis methods were adopted to test the hypotheses. The data analysis process covered four stages: data preprocessing, common method bias test, measurement model evaluation and structural model assessment, all of which were completed with the aid of R 4.4.0 software.

In the data preprocessing stage, descriptive statistical analysis was first conducted on the samples, and the normality (absolute value of skewness < 3, absolute value of kurtosis < 8) and multicollinearity (VIF < 5, tolerance > 0.1) of each observed variable were tested. Subsequently, the Harman single-factor test was used to evaluate the deviation of the common method. A variation of less than 40% explained by the first factor was considered acceptable.

The measurement model was evaluated in terms of reliability and validity. Reliability is judged by Cronbach’s α > 0.7. The convergence validity requirements are that the factor loading is greater than 0.5, the composite reliability (CR) is greater than 0.7, and the average variance extracted (AVE) is greater than 0.5. Discriminant validity was verified by comprehensively using the Fornell-Larcker criterion (where the square root of AVE is greater than the correlation coefficient between variables) and the heterotrait-monotrait (HTMT) ratio < 0.85.

The structural model was evaluated using the bootstrap resampling method (repeated 5,000 times) to estimate the path coefficients and their significance (|z| > 1.96, p < 0.05). The explanatory power of the model was evaluated through R² and the effect size f². The significance of the mediating effect was determined by the bias-corrected bootstrap confidence interval. The moderating effect was revealed by constructing the product term and applying simple slope analysis. The moderated mediating effect was tested by comparing the conditional indirect effects of the moderating variable at different levels.

5 Research results

5.1 Descriptive statistical analysis

5.1.1 Normal distribution test

This study used R 4.4.0 to assess the relations among the measured variables. The sample data conforming to multivariate normality is essential for structural equation modeling; the observed indicators are skewness and kurtosis coefficients. Data are considered normal if the absolute value of skewness is less than 3.0 and absolute value of kurtosis is less than 8.0. According to test results of Table 2, absolute values of the measured items for environmental values (EV), psychological capital (PC), pro-environmental behavior (PEB) and organizational green climate (OGC) range from 0.0014 to 0.2085. The absolute values of skewness for measurement items across the four dimensions range from 0.0014 to 0.2085, and the absolute values of kurtosis range from 0.0217 to 0.5313. All indices satisfy the normality requirements, indicating that the sample data are normally distributed and suitable for further analysis.

Table 2. Descriptive analysis of variables.

5.1.2 Multicollinearity test

Variance Inflation Factor (VIF) is used to measure the degree of correlation between independent variables in multiple linear regression models. It is the ratio of the estimated variance of a regression coefficient with no linear correlation between predictors to actual variance. Larger VIF values indicate severe multicollinearity. It is generally accepted that if VIF values are between 0 and 10 (and are well below 5), multicollinearity is not problematic. In order to detect multicollinearity in the model before model assessment and hypothesis testing, this study used VIF and tolerance to evaluate correlations between measurement items across dimensions.

As shown in Table 3, the VIF values of 20 measures in the four dimensions of environmental values (EV1–EV5), psychological capital (PC1–PC6), pro-environmental behavior (PEB1–PEB4) and organizational green climate (OGC1–OGC5) are 1.7054 to 2.2079 well below the critical value 5 (or 10). The tolerance values for all items are 0.4529 to 0.5864, well above the threshold 0.1. Because the VIF values for all variables satisfy the criteria of being between 0 and 10 (and are well below 5), we conclude that there is no multicollinearity problem among predictors in the structural model. The results therefore provide a solid foundation for subsequent SEM analysis.

Table 3. Collinear diagnosis.

5.2 Test for common method bias

Common method bias (CMB) is a systematic error that can lead to spurious correlations between independent and dependent variables. Podsakoff et al. noted that if severe CMB is present, the data tends to yield one factor or the majority variance is accounted for by one factor (Podsakoff et al., 2003). Harman’s single-factor test was performed prior to statistical analyses by exploratory factor analysis on the sample data. The results (Table 4) show that exploratory factor analysis extracted 4 factors with eigenvalues greater than 1, and the first factor accounted for 33.88% of variance, which is below 40%. This means that no single factor explained the majority variance and hence severe common method bias is unlikely for the data. The core variables of this study (e.g., environmental values, psychological capital) are relatively stable psychological constructs (environmental values as enduring beliefs and psychological capital as a developable positive psychological state) and are little affected by test conditions (paper-based or online) and procedural anonymity was used to reduce social desire bias. Taken together, these considerations suggest that common method bias does not constitute a primary threat to this study.

Table 4. Total variance interpretation.

5.3 Measurement model test

5.3.1 Reliability and validity of measurement model

Reliability was assessed with Cronbach’s α, and convergent validity was evaluated with factor loadings, composite reliability (CR), and average variance extracted (AVE). The criteria are as follows: Cronbach’s α>0.9 indicates good reliability, 0.7–0.8 indicates good reliability and 0.7 is acceptable. Factor loadings >0.7 indicate that the items are significant representations of their latent constructs (a factor loading of 0.50 is acceptable, and loadings <0.4 indicates weak relationship of the item and factor. Composite reliability (CR) is interpreted similarly to Cronbach’s α, with values above 0.7 indicating adequate reliability.

As shown in Table 5, the factor loadings of all items in this study range from 0.695 to 0.811, all above acceptable 0.50 and most above the stricter 0.70 criteria; Cronbach’s α values range from 0.842 to 0.879, all above 0.8, indicating good internal consistency; all AVE values meet or exceed the acceptable threshold of 0.50, ranging from 0.549 to 0.577; CR values range from 0.843 to 0.880, all above the 0.7 threshold, indicating that all dimensions are reliable. Therefore, all variables in this study meet the requirements for validity and reliability.

Table 5. Internal consistency Reliability and convergence effectiveness (Loadings/Composite Reliability/AVE).

5.3.2 Fornell-Larcker criterion for discriminant validity

This study used two methods to evaluate the discriminant validity of the constructs. First, the Fornell-Larcker criterion was applied (Fornell and Larcker, 1981). Fornell-Larcker criterion requires that square root of the construct AVE exceeds correlations with other constructs (Fornell and Larcker, 1981). As shown in Table 6, according to the Fornell-Larcker criterion, the diagonal entries represent the square roots of the AVEs, and the lower-triangle entries represent the inter-construct correlation coefficients. From the table, square roots of the average variance extracted for Environmental Values (EV), Psychological Capital (PC), Pro-environment Behavior (PEB) and Organizational Green Climate (OGC) are 0.765, 0.740, 0.757 and 0.750, respectively, all of which exceed correlations between constructs (larger correlation is 0.522, which is smaller than all AVE square roots).

Table 6. Fornell-Larcker criterion.

Second, the heterotrait–monotrait ratio of correlations (HTMT) was used to assess the discriminant validity of the scales. This method evaluates the mean correlations between indicators of different constructs relative to the mean correlations of indicators within the same construct; an HTMT value below 0.85 (the most stringent criterion) indicates adequate discriminant validity. As shown in Table 7, the HTMT values among constructs in this study ranged from 0.274 to 0.522, with a maximum of 0.522, all below the 0.85 threshold, indicating sufficient discriminant distinctiveness among variables. Thus, discriminant validity of the study variables is supported by this second method.

Table 7. HTMT test results.

5.3.3 Assessment of model explanatory power and effect analysis

The structural equation model results (see Table 8) suggest that the predictors provide moderate explanations of psychological capital and pro-environmental behavior. Environmental values, organizational green climate and their interaction together explain 30.8% of variance in psychological capital (R² = 0.308, p < 0.001), psychological capital, organizational green climate, interaction and environmental values together explain 37.1% of variance in pro-environmental behavior (R² = 0.371, p < 0.001).

Table 8. Explanatory power and effect sizes of predictors.

Further analyses of the effects sizes (f²) of predictors reveal that in psychological capital model environmental values exert a large effect (f² = 0.301) and are the main driver of psychological capital formation. Organizational green climate (f² = 0.127) and interaction effect with environmental values (f² = 0.149) exhibit effect sizes close to the medium range. Organizational green climate contributes not only as main effect but also substantively via moderating relationship between environmental values and psychological capital. In the pro-environmental behavior model psychological capital has the largest effect size (f² = 0.227), showing that it plays a mediating role. Direct effect of environmental values (f² = 0.171), main effect of organizational green climate (f² = 0.142), and interaction effect between psychological capital and organizational green climate (f² = 0.145) are close to medium effect levels, showing that the proposed structural model has good explanation power and that the estimated paths are of substantive theoretical significance.

5.4 Hypothesis testing results

5.4.1 Direct effect tests

In structural equation modeling, path coefficients refer to the regression coefficients between connected variables and measure the influence among variables. Path coefficients were estimated using bootstrapping method, number of replicates of 5000 and significance values greater than 1.96 at significance level 0.05. Significance tests of the path coefficients in the present study yielded the path coefficients together with their corresponding z-values and significance levels; the results are presented in Table 9; Figure 2.

Table 9. Direct test results.

According to the results in Table 9, the path coefficient of environmental value on pro-environmental behavior is 0.237 (z = 3.936 > 1.96, p < 0.001), indicating a significant positive effect of environmental value on employees’ pro-environmental behavior; thus, research hypothesis H1 is supported. The path coefficient from environmental values to psychological capital is 0.443, z = 7.357 > 1.96, p < 0.001, indicating that environmental values have a significant positive effect on employees’ psychological capital; thus, research hypothesis H2 is supported. The path coefficient of psychological capital on employee pro-environmental behavior is 0.346 (z = 5.929, p < 0.001), significant at the 0.01 level. Psychological capital has a positive effect on employee pro-environmental behavior. Thus, hypothesis H3 is supported.

5.4.2 Mediation effect tests

According to the results in Table 10, psychological capital plays a role in mediating environmental values/pro-environmental behavior. The indirect effect of environmental values on pro-environmental behavior via psychological capital is 0.153 (p < 0.001), significant at the 0.01 level. While the direct effect is still significant (β= 0.237, p < 0.001), psychological capital plays a role as a partial mediator between environmental values/pro-environmental behavior, and the mediation effect represents 39.2% of effect. Hypothesis H4 is supported.

Table 10. Mediation effect test results.

5.4.3 Moderation effect tests

As shown in Table 11, organizational green climate significantly and positively modifies the relationship between environmental values and psychological capital (β= 0.205, p < 0.001). Simple slope analysis revealed that when organizational green climate is high, the positive effect of environmental values on psychological capital is stronger (β= 0.648, p < 0.001) and when organizational green climate is low, this effect is weaker and only marginally significant (β = 0.238, p = 0.050). Hypothesis H5 is supported. Organizational green climate significantly moderates the relationship between psychological capital and pro-environmental behavior (β= 0.157, p < 0.001). Simple slope analysis (see Table 12) indicates that when organizational green climate is high, the effect of psychological capital on pro-environmental behavior is significant (β = 0.503, p < 0.001); when it is low, the effect is not significant (β = 0.190, p = 0.067). Hypothesis H6 is supported.

Table 11. Moderation effects test results.

Table 12. Simple slope analysis for moderation effects.

To test this effect, this study used simple slope analysis to test whether organizational green climate moderates the relationship between the positive environmental values (EV) and the positive psychological capital (PC). As shown in Figure 3, PC increases with increasing EV, but the slope differs greatly between levels of OGC.

The simple slope analysis (see Table 12; Figure 3) shows that when OGC is low (-1 SD), positive effect of EV on PC is weak (β= 0.238, p = 0.050); when OGC is high (+1 SD), positive effect of EV on PC is stronger (β= 0.648, p < 0.001). These results suggest that organizational green climate significantly increases positive effect of environmental values on psychological capital, so that in organizations with strong green climate, employees’ environmental values are more effectively converted into psychological capital.

As shown in Figure 4, the positive effect of psychological capital on pro-environmental behavior is also moderated by organizational green climate. Simple slope analysis shows that at low levels of organizational green climate (-1 SD) psychological capital does not have a significant positive effect on pro-environmental behavior (β= 0.190, p = 0.067); at high levels of organizational green climate (+1 SD), psychological capital has a positive effect on pro-environmental behavior (β= 0.503, p < 0.001).

These results suggest that organizational green climate is necessary to convert psychological capital into pro-environmental behavior. When an organization lacks a green climate, even employees with relatively high psychological capital are unlikely to engage in pro-environmental behavior; conversely, when the organization possesses a strong green climate, the promotive effect of psychological capital on pro-environmental behavior can be fully realized.

5.4.4 Moderated mediation effect test

According to the results in Table 13, the indirect effect of environmental values on pro-environmental behavior via psychological capital is moderated by organizational green climate. At low levels of organizational green climate, the indirect effect is nonsignificant (β = 0.045, p = 0.265); at high levels of organizational green climate, the indirect effect is significant and stronger (β = 0.326, p < 0.001); the difference between high and low groups is significant (β = 0.281, p < 0.001). Hypothesis H7 is supported.

Table 13. Conditional indirect effects (moderated mediation).

6 Discussion

6.1 Key findings

Based on a moderated mediation model and 463 shipboard and shore-based employees surveying the path from environmental values to pro-environmental behavior and boundary conditions, we summarized the following results:

First, environmental values predict pro-environmental behavior of workers directly and indirectly by enhancing psychological capital of employees. This shows that psychological capital plays an important role mediating the “awareness-to-action” transition path. This mediating path explained a significant proportion of the variance in the overall relationship between values and behavior.

Second, organizational green climate plays a critical role in this transition path. It both strengthens the positive shaping effect of environmental values on psychological capital and improves the translation efficiency of psychological capital into specific environmental behaviors, positively moderating the entire “values→psychological capital→behavior” mediation path. This shows that strong pro-environmental behavior is a result of the combination of internal psychological resources and a supportive external environment.

Third, from an industry perspective, the structural differences between shipboard and shore-based work contexts may, in theory, lead to differential moderating effects of organizational green climate. However, as job type was only controlled for rather than examined as a moderator in this study, these potential differences remain speculative. We therefore encourage future research to employ multi-group structural equation modeling to test whether the hypothesized pathways operate differently across shipboard and shore-based employees.

6.2 Discussion of results

6.2.1 The mediating role of psychological capital in the awareness-action transformation

This study shows that psychological capital plays a significant role partly in mediating between environmental values and pro-environmental behavior. This is a new theoretical approach to the classic problem of the value-action gap, from the perspective of the Theory of Planned Behavior (Ajzen, 1991), in which answers are sought at cognitive level (say attitudes, subjective norms), to motivational and capability aspects of behavior execution.

According to the conservation of resources theory pro-environmental behaviors require employees to spend more time and effort to respond to uncertainty and setbacks (Hobfoll, 2001). The result is that employees who possess high psychological capital and positive resources view environmental challenges as investments rather than losses. Self-efficacy gives the employees confidence to face challenges, lowering the perceived threshold for action; hope allows them to think strategically about ways to make sense when faced with challenges; resilience guarantees that they can recover quickly from setbacks, preventing a “one failure, total abandonment” phenomenon; optimism maintains expectation of long-term positive results and increasing persistence of action. Psychological capital is the internal motivational and capability engine that transform environmental willingness into concrete, sustained actions.

In high pressure shipping, the difficulties encountered by seafarers with respect to their environment are severe. For example, routes optimized to reduce fuel consumption may be problematic and strict waste sorting is more difficult in the limited space onboard. The positive psychological resources of psychological capital are irreplaceable for maintaining seafarers’ environmental efforts.

6.2.2 The synergistic empowerment mechanism of organizational green climate

One central finding is that organizational green climate not only promotes pro-environmental behavior but also influences it through its interaction with psychological capital in moderated mediation, deepening our understanding of the triadic reciprocal determinism of “person–behavior–environment” in social cognitive theory. It shows that positive psychological resources (psychological capital) of individuals are not static predictors; the effectiveness depends heavily on the support and shaping of the organizational environment (green climate) (Luthans et al., 2007). The study concretizes this dynamic interactional view in environmental management by showing that the organization can nurture individuals’ psychological resources and unleash the potential of behavior.

Organizational green climate provides synergistic empowerment through two ways: First, it acts as an agent for value endorsement: Employees feel that the organization values environmental protection and invests heavily in it, thereby validating their personal environmental values externally. This increased sense of identification activates hope (perceive goals as achievable), optimism (expected positive returns) and efficacy (believing the organization will support). Second, it acts as an instrument for behavior translation: with clear norms, tools, convenient infrastructure and consistent incentives, a green environment reduces situational friction in translating psychological capital into concrete actions, and employees’ confidence and resilience can be translated smoothly without excessive psychological resources to overcome organizational-level barriers.

From a theoretical and industry perspective, it is plausible that this synergistic effect may be especially prominent among crew members, although this possibility awaits empirical verification. Crew members work in a closed “total institution”, have weak direct connections with the organization, depend heavily on company-allocated resources, and face strict safety constraints despite having considerable operational autonomy. In this context, green support signals from the organization, conveyed through material provisions, shipboard management, and company policies, constitute the primary, and sometimes the only, channel for interpreting organizational expectations, obtaining action resources, and sustaining behavioral motivation in an isolated environment. The situational strength of these signals is therefore extremely high. In an open and resource-diverse environment, shore employees are relatively less dependent on the single atmosphere of the organization. The transformation of their psychological capital into behavior may also be more dependent on other situational factors such as team culture or peer pressure.

6.3 Theoretical contributions

Based on the discussion above, the main theoretical contribution of this study is threefold: First, psychological capital from positive psychology is introduced as a main explanation mechanism for environmental management research in terms of intrinsic motivation and volitional processes linking environmental awareness to pro-environmental action. This is different from the traditional rational behavior model based on “cold cognition” prior to decision making. It also uses hot cognition and executive functioning perspectives, resulting in a richer and more practice-oriented explanation of why knowledge does not translate easily into action. Second, it finds the synergistic role of individuals and organizations in generating pro-environmental behavior. By constructing and validating a moderated mediation model, we show that individual psychological capital and organizational green climate do not have independent additive effects on behavior but act together in a “nurture–activate–catalyze” synergy. The model shows that the quality of the organization determines how much personal psychological potential can be activated and converted into actual performance. This provides a precedent for future work to apply and define boundary conditions of relevant theories in a critical but understudied industry. By focusing on the shipping industry (which is historically neglected for organizational behavior research since it has a major environmental responsibility) and through a systematic theoretical analysis of the structural differences between shipboard and shore-based occupational groups, we reveal the contextual boundaries of a general theoretical model when applied to these two distinct work groups.

6.4 Implications

This study provides a systematic and concrete plan for shipping companies to encourage employees to engage in pro-environmental behavior and improve environmental management performance and recommends implementation of a “trinity” strategy:

First, value leadership entails embedding environmental “genes” at the source. Organizations may consider incorporating environmental value traits as soft selection criteria during recruitment. In training systems, efforts should move beyond compliance-based instruction to deepen employees’ personal and identity-based commitment to environmental values through leadership narratives, role modeling, and participation in corporate social responsibility initiatives.

Second, psychological capital development serves to empower employees for environmental action. Firms should integrate psychological capital cultivation into management practice through targeted interventions such as simulation exercises to enhance seafarers’ self-efficacy in handling environmental emergencies, stress management workshops to build resilience, and goal-setting with reward mechanisms to foster hope around concrete environmental objectives. For seafarers in particular, routine at-sea psychological support is especially important.

Third, organizational climate construction requires building a supportive environmental “ecosystem” that combines both hard and soft elements. At the hardware and institutional level, vessels and offices should be equipped with adequate environmental facilities and their usability optimized. A transparent and quantifiable environmental performance appraisal system should be established and linked to promotion and compensation for masters, chief engineers, and shore-based managers. At the leadership and cultural level, green transformational leadership should be promoted among managers at all levels, including vessel leaders, so that role modeling, vision-based motivation, and individualized consideration collectively convey the organization’s commitment to environmental protection.

By using the above integrated strategies, shipping companies can turn individual employees’ environmental awareness into robust and sustainable organizational green competitiveness and gain a strategic advantage in a competitive market that demands sustainability.

7 Conclusions

7.1 Research conclusions

This study explains how environmental values are translated into pro-environmental actions. It is found that employees’ environmental values are most effectively translated into sustained pro-environmental actions when they stimulate psychological capital as a vital internal resource and work in concert with a supportive organizational environment. The theoretical contribution of this “individual–context synergy model” lies in the possibility that environmental management research extends beyond technical and regulatory perspectives to a more detailed analysis of the interaction between employee psychological states and organizational environment, providing a new theoretical framework for understanding and managing environmental behavior in organizations.

7.2 Limitations and remarks

This paper has the following limitations: First, although the cross-sectional design, supported by theoretical reasoning and statistical methods, shows relationships between variables, it is hard to establish causal ordering. Successful experiences of pro-environmental behavior may reciprocally influence psychological capital; this dynamic reciprocal relationship needs longitudinal research. Second, self-report data, derived from a single source, may be subject to social desirability biases and common method variance. Although statistical tests show common method bias is not harmful, future studies that incorporate objective indicators (e.g. energy consumption data) or multi-source evaluations would increase robustness of results. Moreover, the contextual relevance of the measurement scales could be further improved. Specific work contexts in shipping industry (e.g. decision-making pressures under extreme sea conditions) may require more industry specific measurement instruments to capture real psychological state and behavior of employees. In this study, seafaring versus shore-based occupational status was treated solely as a control variable, and its potential systematic moderating role within the overall model was not examined in depth. Future research should employ larger samples and use multi-group structural equation modeling to formally test whether shipboard versus shore work contexts moderate the proposed mediation pathways, as the current study did not examine this question empirically.

7.3 Future research directions

Based on these results and limitations, future research could be pursued in the following directions: (1) conducting diary studies or experience sampling to monitor daily fluctuations in employees’ environmental motivations and behaviors; (2) integrating multi-level factors such as team climate and leadership styles to build a more systematic explanatory framework; (3) examining the double-edged sword effects of environmental pressure, such as anxiety and burnout; and (4) conducting intervention studies to test the practical effectiveness of measures such as psychological capital training and organizational climate initiatives.

The final conclusion of this study is that building a sustainable future demands not only technological innovation and improvement of institutions but also a culture of organization that motivates and sustains individual environmental action. Only when institutional design and humanistic care complement each other, the willingness of employees to act can translate into lasting action, achieving a balance between economic development and environmental protection.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. Generative artificial intelligence was employed for grammar checking and language refinement of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Yi-Che Shih, National Cheng Kung University, Taiwan

Reviewed by: Xinjian Wang, Dalian Maritime University, China

Botang Li, Guangzhou Maritime College, China