2.1 Pro-Environmental Behavior at Work (PEWB)

Pro-Environmental Behavior at Work (PEWB) refers to employees’ actions aimed at reducing environmental harm in the workplace, whether derived from organizational activities or individual conduct (Alherimi, 2024; Sierra-Barón & Meneses Báez, 2018). These behaviors can be task-related or proactive, extending beyond formal responsibilities (Bissing-Olson et al., 2013). PEWB includes activities such as recycling, pollution prevention, and promoting sustainability among colleagues (Foster et al., 2022; Ones & Dilchert, 2012b).

Promoting environmentally friendly actions not only supports ecological preservation but also enhances organizational performance (Alherimi et al., 2024; Mouro & Duarte 2021). The concept of PEWB has gained academic importance by highlighting the crucial role of human behavior in achieving environmental sustainability. It is associated with various forms of pro-environmental conduct, including responsible environmental behavior, sustainable environmental behavior, impactful environmental behavior, green behavior, ecological actions, and environmentally friendly conduct (Omarova & Jo, 2022).

However, despite its theoretical and practical contributions, the conceptualization of PEWB continues to present challenges. Some scholars regard it as an extension of voluntary organizational behavior, while others view it as an autonomous practice, independent of organizational goals. This divergence highlights the conceptual heterogeneity that characterizes the existing literature. For example, Wesselink et al. (2017) outline three research strands: (1) Organizational Citizenship Behavior for the Environment (OCB-E) grounded in voluntary cooperation (Boiral, 2009; Lamm et al., 2015), (2) personal and social determinants based on Kollmuss and Agyeman’s (2002) framework, and (3) Employee Green Behavior (EGB), which differentiates between proactive and task-based behaviors (Norton et al., 2014; Ramus & Killmer, 2007).

Recent literature reviews have expanded this field. Zaidi and Azmi (2024) identified six thematic clusters, including green human resource management (HRM) and sustainable consumption, while Widyanty et al. (2025) grouped research into organizational, individual, and HRM factors. Despite the development of several explanatory models, there has been limited empirical validation. (Blok et al., 2015; McDonald, 2014; Young et al., 2015). Moreover, a critical gap persists regarding the relevance and applicability of these models in Latin American and low-income contexts (Ciocirlan, 2016).

In summary, PEWB is essential for fostering organizational sustainability. While its determinants and expressions are increasingly understood, the scarcity of empirical studies, especially in Latin America, underscores the need for more contextually grounded research to enhance both theoretical models and practical implementation.

2.2 Determinants of Pro-Environmental Behavior at Work (PEWB)

The predictive model developed by Blok et al. (2015) offers a comprehensive framework for explaining pro-environmental workplace behavior (PEWB). This model integrates both individual (internal) and organizational (external) factors, positioning Intention to Act (IA) as a key mediating variable between psychological and contextual determinants and observable behavior, in line with the Theory of Planned Behavior (TPB) (Ajzen, 1991).

The model’s theoretical structure draws from three complementary frameworks. First, TPB states that behavior is guided by intention, which is influenced by attitudes, subjective norms, and perceived behavioral control (PBC). Second, Norm Activation Theory (NAT) holds that personal norms, moral obligations, are activated when individuals recognize the consequences of their actions and feel personally responsible (Schwartz, 1977, 1994). Third, the Value-Belief-Norm (VBN) theory suggests that personal values, particularly altruistic ones, shape an ecological worldview that activates moral norms, thus motivating pro-environmental behavior (Stern et al, 1999, Stern 2000).

Within this framework, the model highlights several key internal factors. Personal norms (PN) are among the strongest predictors of both IA and PEWB (Blok et al., 2015; Niu et al., 2023; Ruepert et al., 2016; Stern et al., 1999, Stern, 2000Zibarras et al., 2025), with their impact strengthened when contextual variables are present (Yuriev et al., 2020a). Social norms (SN), or shared expectations in the workplace, also positively influence both intention and behavior by promoting behaviors like recycling and energy saving (Blok et al., 2015; Fikria et al., 2024; Gao et al., 2017; Stern et al., 1999; Stern, 2000; Wu et al., 2024). Perceived behavioral control (PBC), reflecting employees’ perceived ability to act pro-environmentally, is a consistent predictor of both IA and PEWB (Ajzen, 1991; Wesselink et al., 2017).

Another important factor is environmental awareness (EA), defined as knowledge of the consequences of human behavior on the environment (Grob, 1995). EA is positively associated with both intention and behavior (Blok et al., 2015; Cao et al., 2024; Fu et al., 2020; Norton, 2016). Attitudes toward the environment (ATE), reflecting evaluative beliefs about environmental actions, also predict IA and PEWB (Ajzen & Fishbein, 2004; Blok et al., 2015; Fishbein & Azjen 1975; Norton, 2016). In addition, the need for information (NI) about environmental policies and initiatives indicates employees’ curiosity and engagement, serving as a motivational factor for PEWB (Blok et al., 2015; Bertilsson & Remle, 2018).

From the VBN perspective, general values (GV), such as openness to change, altruism, and self-transcendence, significantly influence PN and PEWB (Costa et al., 2022; De Groot & Steg, 2008; Blok et al., 2015; Luís & Silva, 2022; Paillé et al., 2020;). Similarly, environmental values (e.g., biospheric values) reinforce moral norms and intentions (Ruepert et al., 2016), complementing the influence of EA and GV.

The model also incorporates organizational-level (external factors), which, although not central to TPB, align with its constructs. Situational factors (SF) such as access to green infrastructure, facilitate pro-environmental behavior (Blok et al., 2015; Costa et al., 2022; Gusmerotti et al., 2023; Leitão et al., 2024; Ranjan et al., 2021). Exemplary leadership encourages pro-environmental norms and strengthens IA and PEWB (Albrecht et al., 2024; Blok et al., 2015; Gusmerotti et al., 2023; Zhao et al., 2023; Xu et al., 2022), while leadership support exerts a direct influence on PEWB (Blok et al., 2015; Costa et al., 2022; Tosti-Kharas et al., 2016; Luís & Silva, 2022; Paillé et al., 2020).

Organizations aiming to promote sustainability increasingly recognize the importance of employees’ engagement in pro-environmental workplace behavior (PEWB). Previous studies highlight that both personal motivation (internal factors) and organizational context (external factors) significantly shape how employees perceive and engage in ecological behaviors at work (Robertson & Barling, 2015; Norton et al., 2015).

Based on this framework, this study classifies the predictors into internal factors (e.g., personal norms, environmental awareness, values) and external factors (e.g., situational factors, exemplary leadership behavior, leader support) to evaluate their direct and indirect effects (via Intention to Act [IA]) on PEWB.

Internal factors such as personal norms, social norms, perceived behavioral control, environmental awareness, attitudes toward the environment, general values (Altruistic values, conservative values, Enhancement values, and openness to Change values), and the need for information on environmental initiatives are expected to directly influence employees' PEWB. Additionally, external factors including situational factors, exemplary leadership behavior, and leader support are hypothesized to have a direct influence on PEWB. Furthermore, intention to act (IA) is posited as a direct predictor of PEWB.

3.1 Study

This cross-sectional study examined a model of relationships between personal and organizational employee variables, based on Block's theory (Block et al., 2015) about internal and external factors in PEWB (see section 2.3).

3.2 Participants

The study included a convenience sample of 856 employees from various organizations in the Huila department, all of whom voluntarily provided informed consent. The sample comprised 51.4% women (n = 414) and 48.6% men (n = 416), with a mean age of 35.8 years (SD = 12.7).

Regarding marital status, 63.7% of participants reported having a partner, 25.1% identified as single, 6.4% as divorced, and 4.8% as widowed. In terms of educational attainment, 54% held university or postgraduate degrees, 37.6% had completed technical or technological education, 4.4% had finished high school, and 3% had primary-level education.

The socioeconomic analysis revealed that 75.9% of participants belonged to strata 2–3, with the remaining individuals classified in strata 4 or higher. Concerning employment status, 53.7% held permanent or fixed-term contracts, 23.8% worked under service agreements, 13.3% were engaged in apprenticeships, and 6.3% occupied temporary positions.

Participants reported diverse organizational roles: 62% worked in administrative positions (including support, clerical, specialized, and secretarial roles), 23% served as teachers, instructors, or consultants, 7% held department head roles, and 5% worked in sales.

3.3 Instruments

This study employed scales adapted from Blok et al. (2015) and contextualized by Sierra-Barón and Meneses Báez (2022b) to measure the internal and external factors influencing Pro-Environmental Behavior at Work (PEWB) among employees in the Huila region, Colombia. These factors and their corresponding measurement scales are described as follows:

3.4 Procedure

The researchers conducted data collection online. Initially, they emailed participants an informed consent form, followed by sociodemographic questionnaires and instruments designed to measure internal and external factors related to pro-environmental behavior, intention to act, and the Pro-Environmental Behavior at Work (PEWB) Questionnaire. All instruments had been previously validated for the Colombian population.

To identify potential participants, the research team contacted the Neiva Chamber of Commerce. The Chamber provided a list of email addresses and phone numbers from organizations located within the Huila department. Using this information, the team extended invitations to these organizations. Participating organizations then received the questionnaires in digital format. In some instances, they also shared internal employee contact lists to facilitate the direct distribution of the instruments.

Eligibility for participation required individuals to be of legal age, to be employed on-site at an organization within the Huila department, and to hold a position under a legally recognized employment contract in Colombia.

Data Analysis. The data analysis followed a five-step procedure using SPSS v.25.0 and AMOS v.25, in alignment with methodological standards in behavioral research.

Step 1: Preliminary screening

Researchers calculated item- and scale-level non-response rates and conducted descriptive statistics for all sociodemographic variables and for the scales measuring internal factors, external factors, and Pro-Environmental Behavior at Work (PEWB).

Step 2: Testing for normality and correlations

The Kolmogorov–Smirnov test indicated that most variables did not follow a normal distribution (p > .05). Accordingly, Spearman’s rank-order correlation was used to examine bivariate relationships between internal/external factors, Intention to Act (IA), and PEWB. Correlations were assessed for statistical significance (p ≤ .05) and to ensure the absence of multicollinearity (r < .85), following the criteria proposed by Howitt and Cramer (2017) and Pui-Wa and Qiong (2007).

Step 3: Confirmatory Factor Analysis (CFA)

To evaluate the conceptual validity of the measurement instruments, a CFA was performed using AMOS v.25. Model fit was assessed through the following goodness-of-fit indices: Chi-square ( ${\chi }^2$): p < .05 (though sensitive to sample size); Comparative Fit Index (CFI): > .95 indicates excellent fit Tucker–Lewis Index (TLI) and Normed Fit Index (NFI): > .90 acceptable, > .95 excellent Adjusted Goodness-of-Fit Index (AGFI): > .80 acceptable Root Mean Square Error of Approximation (RMSEA): < .05 excellent; .05–.08 acceptable (Hair et al., 2020; Collier, 2020)

Step 4: Reliability Analysis

Internal consistency for each scale was evaluated using both Cronbach’s alpha and McDonald’s (2014) omega coefficients, including 95% confidence intervals, estimated through the macro developed by Hayes and Coutts (2020).

Step 5: Structural Equation Modeling (SEM) and Mediation Testing

To test the theoretical model proposed by Blok et al. (2015) —which posits that internal and external factors influence Pro-Environmental Workplace Behavior (PEWB) through Intention to Act (IA)— a Structural Equation Model (SEM) was developed and estimated using AMOS v.25 (Arbuckle, 2017), following recommendations by Hair et al. (2020) and Iacobucci (2008). The model incorporated both latent and observed variables and specified direct paths from internal and external predictors to PEWB, as well as indirect paths mediated by IA.

Estimation was conducted using the maximum likelihood (ML) method. The model was graphically specified using AMOS’s path diagram interface. Based on theoretical and conceptual overlap, regression weights from Personal Norms (PN) and Social Norms (SN) to PEWB were constrained to equality, reflecting their shared role as moral constructs (Schwartz, 1977; Blok et al., 2015). This constraint also helped reduce the number of free parameters (n = 105), thereby improving model parsimony.

To test for mediation, indirect effects were analyzed using bootstrapping methods, consistent with contemporary mediation analysis procedures (e.g., Rozeboom, 1956; Ruiz et al., 2010). Specifically, the significance of indirect paths from internal and external variables to PEWB via IA was evaluated.

Model fit was assessed using standard goodness-of-fit indices, with the following criteria:

This analytical approach allowed for empirical validation of the theoretical model and shed light on the mechanisms by which internal and external factors affect PEWB—both directly and indirectly—through the mediating role of IA, aligning with established principles of mediation theory.

4.1 Descriptive statistics study variables

The descriptive statistics for the scales measuring internal factors, external factors, Intention to Act (IA), and PEBW are displayed in table 1. Among the internal factors, the Environmental Values Scale (EVS) had the highest mean score (M = 9.98, SD = 3.50), followed by Personal Norms (PN; M = 12.11, SD = 3.87). For external factors, the Leader Support (LS) scale recorded the highest average (M = 14.06, SD = 4.58). The mean score for PEBW was 66.36 (SD = 18.75), indicating a moderately high level of perceived pro-environmental behavior in the workplace.

4.2 Correlation coefficients between study variables

The Spearman correlation coefficients among all study variables are shown in Table 2. As expected, PEBW presents strong and statistically significant positive correlations with key internal factors such as Environmental Awareness (EA; ρ = .82, p < .01), Personal Norms (PN; ρ = .78, p < .01), and Altruistic Values (AV; ρ = .72, p < .01). External factors also demonstrated significant associations, notably with Leader Support (LS; ρ = .65, p < .01) and Environmental Leadership Behaviors (ELB; ρ = .79, p < .01). Intention to Act (IA) was also strongly correlated with PEWB (ρ = .80, p < .01), reinforcing its mediating role between motivational factors and actual behavior.

4.3 Confirmatory Factor Analysis (CFA)

We conducted a confirmatory factor analysis (CFA) using AMOS v.25 (Arbuckle, 2017) to evaluate the factor structure of internal factors, external factors, Intention to Act (IA), and Pro-Environmental Behavior at Work (PEBW), based on the model proposed by Blok et al. (2015).

The internal factors scale, consisting of 11 subscales and a total of 50 items, showed a good model fit ( ${\chi }^2$ = 2440.208; df = 1259; p = .000; CFI = .943; TLI = .937; NFI = .889; RMSEA = .033), in accordance with Hair et al.’s (2020) criteria for acceptable model fit.

The external factors scale, which included 3 subscales and 11 items, also demonstrated a strong model fit ( ${\chi }^2$ = 93.177; df = 41; p = .000; CFI = .982; TLI = .976; NFI = .969; RMSEA = .039).

The Intention to Act (IA) scale, composed of 3 items, achieved an excellent model fit ( ${\chi }^2$ = .102; df = 1; p = .75; CFI = 1.000; TLI = 1.005; NFI = 1.000; RMSEA = .000), meeting the standards recommended by Hair et al. (2020).

Lastly, the PEWB scale, which included 28 items, showed satisfactory model fit indices ( ${\chi }^2$ = 1072.669; df = 350; p = .000; CFI = .919; TLI = .913; NFI = .885; RMSEA = .049), also in line with Hair et al. (2020).

4.4 Reliability Analysis of the Scales

In general, the Colombian scales used in this study (table 3) demonstrated adequate internal consistency (α ≥ .70), with the exception of the scales for Perceived Behavioral Control (α = .55), Situational Factors (α = .65), and Environmental Leadership Behaviors (α = .67), which showed lower reliability coefficients.

As presented in Table 3, the internal consistency coefficients obtained from the Colombian version are generally slightly lower than those reported for the Dutch version (α ≥ .70). An exception to this trend is the Perceived Behavioral Control scale, which demonstrated a higher reliability coefficient in the Colombian sample (α = .55) compared to the Dutch sample (α = .40).

4.5 Mediation model

We evaluated the model proposed by Blok et al. (2015), which suggests that both internal factors —including Personal Norms (PN), Social Norms (SN), Perceived Behavioral Control (PBC), Environmental Awareness (EA), Altruistic Values (AV), Conservative Values (CV), Enhancement Values (VAM), Openness to Change Values (OCV); Environmental Values (EV), Attitude toward Environmental Behavior (AEB), and Information Need (NI)— and external factors —situational factors (SF), Exemplary leader behavior (ELB), and leader support (LS)— exert direct and indirect effects on Pro-Environmental Behavior at Work (PEWB), with Intention to Act (IA) serving as a mediating variable. Using structural equation modeling (SEM), we examined both the direct effects and those mediated by IA on PEBW.

Given the high number of free parameters to estimate in SEM and the complexity of the variance-covariance matrix, the initial degrees of freedom resulted in zero, producing a just-identified model. To address this issue, we constrained the regression weight of Personal Norms (PN) and Social Norms (SN) on Pro-Environmental Behavior at Work (PEWB). This decision was guided by two main reasons: first, the freely estimated weights for these variables showed close and significant effects on PEWB; second, conceptually, both variables have similar direct effects on PEWB, as suggested by Blok et al. (2015). The constrained model demonstrated satisfactory fit indices ( ${\chi }^2$ = 1.328, df = 1, p = .249; CFI = .999; TLI = .997; NFI = 1.000; RMSEA = .020), indicating a good fit of the model to the data. Figure 1 presents the final model adopted for this study.

Figure 1Model of internal and external factors mediated by IA on PEBW 

The results of the structural equation modeling (SEM) analysis showed the direct effects of the model's variables on Pro-Environmental Behavior at Work (PEWB). Among the internal factors, Personal Norms (PN), Social Norms (SN), Environmental Awareness (EA), Enhancement Values (VAM), Environmental Values (EVS), Attitudes Toward the Environment (AEB), and Need for Information (NI) showed significant and positive direct effects on PEWB (p < .05). Similarly, among the external factors, Leadership Support (LS) and Exemplary Leader Behavior (ELB) also demonstrated positive direct effects on pro-environmental behavior (Table 4). Additionally, Intention to Act (IA) shows a significant direct effect on PEBW. These findings confirm that both personal and contextual factors directly influence employees’ pro-environmental behaviors in the workplace, with intention to act working as a mediating variable.

5.1 Practical implications and theoretical contributions

From a practical perspective, this study offers several insights:

Strengthening Behavioral Intentions: Organizations should design interventions to reinforce employees’ intention to act pro-environmentally by aligning internal values with institutional support mechanisms. This includes providing regular training, promoting environmental awareness, and facilitating access to green infrastructure.

Role of Leadership: Visible and supportive leadership is essential for fostering sustainability. Managers should model pro-environmental behaviors and actively encourage participation in sustainability initiatives. Incorporating environmental stewardship into leadership development programs may amplify these efforts.

Value-Based Human Resource Practices: Identifying values such as openness to change and environmental values as predictors of PEWB highlights the need to integrate value-based components into HR practices, including recruitment, evaluation, and reward systems that recognize pro-environmental behaviors.

From a theoretical standpoint, this study contributes to the refinement of the Theory of Planned Behavior by demonstrating the mediating role of Intention to Act across a broader set of predictors. Furthermore, the results support the relevance of the Value-Belief-Norm framework for capturing the interplay between values, intentions, and behaviors in organizational contexts; as well as, highlight the importance of integrating multiple behavioral theories to adequately capture the complexity of environmental behavior in organizational settings.

Overall, this research enhances understanding of how internal motivations and organizational contexts interact to promote sustainable workplace behavior, particularly in underrepresented regions such as Latin America, offering insights applicable to other emerging economies facing similar challenges.

5.2 Limitations

Although this research provides valuable insights into the determinants of Pro-Environmental Behavior at Work (PEBW), several limitations should be considered.

First, the use of a cross-sectional design limits the ability to draw causal inferences. Longitudinal studies would be more effective in capturing changes in employee behavior over time and determining causality.

Second, data were collected using self-reported instruments, which may be subject to social desirability bias. Participants might have overreported their engagement in pro-environmental behavior, potentially skewing the results. Future research could benefit from incorporating observational or supervisor-reported measures to triangulate findings.

Third, the study focused solely on organizations in the Huila Department of Colombia. While this region provides an important context for understanding environmental behavior in Latin America, the findings may not be generalizable to other regions or countries with different cultural, institutional, or economic dynamics.

Finally, some variables showed small effect sizes in indirect pathways, raising the possibility of spurious relationships. Further studies should validate these associations using larger and more diverse samples, and explore additional moderators or mediators such as organizational culture or green identity that may influence PEBW.