Minding the Circularity Gap: Modelling the Impact of Green Organisational Strategy and Norms on Employees’ High-Intensity Green Behavioural Intentions

Employee Green Behavioural Intentions Versus Enacted Employee Green Behaviour

We acknowledge that there is a distinction between behavioural intention and actual behaviour. Intention is a mental state where employees have a plan to do something (e.g., plan for or support the reuse of construction products), which may or may not lead to the employee acting on the intention. Although we do not directly build upon the theory of planned behaviour (Ajzen, 1991), we similarly propose that behavioural intention is an antecedent to behaviour. Some studies have found a strong unidirectional relationship between green intention and green behaviour (e.g., Hasebrook et al., 2022), while others have found a moderate relationship when there is a strong green psychological climate (Norton et al., 2016). The theory of planned behaviour is one of the most widely applied theories to explain employee green behaviour through intentions in an organisational context and has been applied in similar studies that incorporate it with the Norm activation model (NAM) constructs (e.g., Chan et al., 2022; Wall et al., 2007). According to this, intentions may be a proxy that guides actions; however, the theory of planned behaviour has been criticised for the possible intention-behaviour gap (Ajzen, 2020). However, some studies suggest that stronger intentions are stable and increase predictability, in which moderators like a high level of moral norms, such as pro-environmental personal norms, seem to contribute to narrow the gap (e.g., Conner & Norman, 2022).

Moreover, there are numerous conceptualisations and measurement scales of employee green behaviour and intentions across different research fields (for an overview, see, e.g., Francoeur et al., 2019; Katz et al., 2022; Zhang et al., 2024). Although the terms employee green behaviour and high-intensity behavioural intentions are used throughout this paper, we acknowledge that it is closely related to other conceptualisations, such as employee or workplace proenvironmental behaviour (e.g., Paillé, 2018; Robertson & Barling, 2013; Wells et al., 2018a; Young et al., 2015; Zaidi & Azmi, 2024), eco-initiatives and eco-innovation (Ramus, 2001; Ramus & Steger, 2000), organisational citizenship behaviour for the environment (Boiral & Paillé, 2012), microlevel corporate social sustainability behaviour (Aguinis & Glavas, 2012; Glavas, 2016; Glavas & Radic, 2019), environmentally responsible workplace behaviour (Smith & O'Sullivan, 2012), employee sustainability change agency (Russell et al., 2021) and so on.

According to Boiral et al. (2015a), employee green behaviour can therefore be understood as an umbrella term stemming from different research streams. To unify the fragmented research within organisational greening, there is a need to adopt a broad and inclusive understanding of employee green behaviour and intentions (Francoeur et al., 2019). Following this line of reasoning, employee green behavioural intentions can be grouped into five major categories of intentions to engage in positive and negative proenvironmental workplace behaviours: transforming, conserving, avoiding harm, influencing others, or taking initiative (i.e., green-five taxonomy, see Ones & Dilchert, 2012a). Counterproductive sustainability behaviours have later been added to the original green-five taxonomy (Ones et al., 2018). This comprehensive taxonomy has been adopted by several researchers, recent reviews and meta-analyses (Francoeur et al., 2019; Wiernik et al., 2016; Zacher et al., 2022).

Reuse of construction products as a type of employee green behavioural intention could be classified across many of these categories, such as embracing reuse by changing how the work is performed and choosing reuse as an alternative to virgin materials (i.e., transforming) or reducing use of virgin products to prevent pollution (conserving/avoiding harm) and prioritising and supporting projects that include reuse (taking initiative/influencing) (Ones & Dilchert, 2012a). However, in this study, reuse of construction products was conceptualised broadly to match diverse roles and contexts, because it is a form of high-intensity green behavioural intention that depends on the type of company (e.g., consultancy firms, building owners/developers, manufacturers, etc.) and the job role (e.g., architects, engineers, environmental consultants, project leaders, etc.).

Moreover, employee green behavioural intentions can have a direct effect (employees act themselves) or an indirect effect (employees encourage others), be extrarole (organisational citizenship behaviour) or in-role (task-related behaviour) and be low-intensity or high-intensity behaviours (related to degree of effort, risk and change) (Francoeur et al., 2019; Zacher et al., 2022). Reuse of construction products may be regarded as high-intensity behaviour because it is, due to several challenging barriers, not the standard in the current construction industry (Fufa et al., 2023; Knoth et al., 2022); therefore, it involves radical changes, major effort and high risk for both the employee and the organisation. In addition, the reuse of construction products is conceptually different from discrete and low-effort office behaviours. For example, turning off office lights is not related to required job-tasks in the same way as reuse of construction products, because reuse changes the nature of the task; instead of working with a linear flow of construction products in projects (i.e., buy new and dispose old products when building), reuse would entail working with a circular flow of products (i.e., reuse products). This changes the workflow in the entire value chain. Although the environmental management literature has focused more on formal practices and nonvoluntary behaviours, it is often considered from an organisational-level point of view (Boiral et al., 2015b). Because most of the organisational psychology research has focused on voluntary and discrete employee green behaviours, neglecting wider sector-specific behaviours, there is a call for research on contextual behaviours that goes beyond green office behaviours (Francoeur et al., 2019; Norton et al., 2015aa). Therefore, we aim to add to the literature by focusing on antecedents to specific, high-effort employee green behaviours.

Environmental Organisational Strategies

Environmental organisational strategies are formal structures at the organisational level of analysis, such as green policies or goals (Zacher et al., 2022), and are related to corpo/mesolevel factors in a system (Flagstad et al., 2021; Guiyao et al., 2023). Strategies are often formed to either comply with external pressure (reactive) or internal motives (proactive) for a company to become more sustainable, which is different to taking strategies into action with environmental initiatives such as green Human Resources Management (HRM) practices (Dilchert et al., 2018). Green HRM can be understood as different HRM activities, such as employee training or recruitment, which are implemented to encourage employee green behaviour or intentions (e.g., Dumont et al., 2016).

Different disciplines tend to operationalise environmental organisational strategies differently. For example, within the green HRM literature, there has been a large focus on the effect of environmental initiatives (e.g., green training), and within the corporate social responsibility (CSR) field, there has been an increased focus on how micro-CSR strategies and initiatives affect employee-level outcomes (for an overview, see for example Glavas & Radic, 2019; Katz et al., 2022; Zacher et al., 2022). However, there is still a need for more research on the underlying mechanisms of this relationship (Glavas & Radic, 2019). For example, micro-CSR theories are often based on organisational-level antecedents (e.g., if the company has a strategy or not) linked to employee-level outcomes; however, individual perceptions of CSR are crucial because the employees need to be aware of these strategies and might perceive them differently (Glavas & Radic, 2019). The literature suggests that formalised organisational strategies will remain symbolic and insufficient unless the employees engage with them (e.g., Norton et al., 2015a, 2018; Yuriev et al., 2018). Although environmental organisational strategies and CSR are related to different research streams and are conceptualised differently, they have the strategic green top-down aspect in common. Environmental organisational strategies have received less attention in the organisational psychology domain, partly because these types of contextual factors tend to be overlooked (for an exception, see Norton et al., 2014, 2016). However, the literature suggests that measuring both antecedents and outcomes at the same level of analysis would lead to more proximal prediction (Norton et al., 2016). Therefore, further theoretical development is necessary to understand how environmental organisational strategies influence employees and how they are put into practice. In this paper, we apply multiple measures to capture the perceived presence of a context-specific environmental organisational strategy for reuse of construction products (i.e., perceived current policy for reuse, company plans for reuse in the future and ambitions for reuse in specific projects). Hopefully, this will advance the current understanding that is mainly based on single measures of a general sustainability policy (e.g., Norton et al., 2014).

Another argument for studying environmental organisational strategies is the potential negative effects of these strategies. Many firms may be pushed by their stakeholders to position themselves in relation to the growing demand for green products, environmentally friendly services and better environmental performance. In this, there is a risk for greenwashing, which is to make misleading claims regarding, for example, the environmental performance of the firm (Delmas & Burbano, 2011). Currently, the grey area appears to be relatively large in terms of what can be claimed as green. One approach to limit greenwashing is to make use of environmental certifications, which can be compared to environmental organisational strategies because they are more top-down than bottom-up, and mostly originate at the macro level rather than the micro level. Although many positive outcomes have been identified (Boiral et al., 2018), such as corporate environmental performance (Tolosa et al., 2019), it also seems that certification schemes can create cognitive dissonance when there is incongruence between requirements and environmental values (Flagstad et al., 2022). In the latter study, an ‘eco grey zone’ was identified as the area between greening and greenwashing (as outcomes of the certification process). The purported existence of an ‘eco grey zone’ indicates that working towards environmental sustainability within a company can be complicated and sometimes paradoxical, and that a strong green organisational climate (which includes coworkers’ proenvironmental social norms) may help steer a company towards environmental sustainability rather than greenwashing (Flagstad et al., 2022). The literature within organisational psychology has found a positive relationship between the perception of a general environmental organisational strategy and a green psychological climate (Norton et al., 2014, 2016). Thus, we argue that the perceived presence of organisational environmental strategies facilitates employee high-intensity green behavioural intentions though coworker’s proenvironmental social norms in the workplace. We therefore hypothesise that:

The Role of Coworkers’ Proenvironmental Social Norms

Social factors in the workplace include social influence from important others in the organisation (e.g., teams, coworkers, leaders) (Zacher et al., 2022). From a systems perspective, this is related to the employees’ micro or team/leader environment (Flagstad et al., 2021; Guiyao et al., 2023; Zacher et al., 2022). Normative social influence is a common strategy used to explain proenvironmental behaviour and behavioural intentions (Cialdini & Jacobson, 2021; Farrow et al., 2017). Within environmental psychology, it is extensively understood as a social contextual factor influencing general proenvironmental behavioural intentions and actual behaviour (Bamberg et al., 2007). This concept comprises what others think and do and how this may influence individuals to behave in certain ways, for example, environmentally friendly (Schultz et al., 2014; Thøgersen, 2009). Hence, social norms are ‘shared societal expectations that impact on behaviour’ (Reynolds et al., 2015, p. 2). The literature typically distinguishes between two broad types of social norms: injunctive/subjective social norms (i.e., what others think are acceptable ways to behave) and descriptive social norms (i.e., what behaviour others typically engage in) (e.g., Cialdini et al., 1991; Cialdini & Jacobson, 2021; Farrow et al., 2017; Niemiec et al., 2020; Reynolds et al., 2015). Descriptive social norms seem to affect behavioural intentions and behaviour positively and to a greater extent than injunctive norms (de Groot et al., 2021), or they have a similar effect (Helferich et al., 2023; Niemiec et al., 2020). According to Niemiec et al. (2020), fewer studies focus on descriptive norms than on injunctive/subjective norms. This may be due to a widespread application of the theory of planned behaviour (Ajzen, 1991) in predicting proenvironmental behavioural intentions and behaviour, which focuses specifically on injunctive norms. Through this theoretical lens, social norms motivate behavioural intentions through a rational cost-benefit assessment of informal social rewards and sanctions for the behaviour (Guiyao et al., 2023). For example, ‘not suggesting reuse in a construction project is wrong and colleagues will disapprove’. Descriptive social norms also motivate through a cost-benefit assessment, but the focus is on beliefs about common behaviours by a social group. For example, if an employee observes that most colleagues try to reuse construction products in projects, they might start to initiate this too, because ‘everyone is suggesting reuse options in construction projects in this organisation’. This paper focuses specifically on descriptive norms as the perceived presence of coworkers’ involvement in employee green behaviour, in line with Norton’s concept of coworkers’ green psychological climate (i.e., typical behaviour of coworkers) (e.g., Mouro & Duarte, 2021; Norton et al., 2014, 2015a, 2016). In the organisational psychology field, it is well established that different organisational and psychological climates, such as safety climate, influence employee behavioural intentions and actual behaviour (e.g., safety) (Schneider et al., 2013). A green psychological climate has been found to be positively associated with employee green behaviour and behavioural intentions (e.g., Mouro & Duarte, 2021; Norton et al., 2014, 2016; Zafar & Suseno, 2024). We thus hypothesise that:

The Function of Proenvironmental Personal Norms

Personal factors are antecedents to employee green behaviour and behavioural intentions internal to the employee (Zacher et al., 2022). One such factor is personal norms – feelings of moral responsibility to act in a certain manner in a specific context (Schultz et al., 2014). The concept of personal norms originates from the NAM. In this model, the activation of personal norms directly influences behaviour when the person is aware of the consequences and takes personal responsibility for their actions (Schwartz, 1977). People hold personal norms that may or may not become activated, and in the case of activation, generate behaviour. In an organisational context, a green organisational climate and environmental organisational strategies can reinforce psychological mechanisms, such as awareness of consequences and ascription of responsibility, which are central to the NAM. Once these mechanisms have become active, they may shape the motivation to act, thereby increasing the likelihood of proenvironmental intention. Accordingly, the model can indicate how a psychological green climate, environmental organisational strategies and personal norms promote high-intensity green behavioural intentions.

The NAM is prominent in social and environmental psychology, and research has found strong positive links between proenvironmental personal norms and proenvironmental behaviour or behavioural intentions in the private sphere (Helferich et al., 2023). Organisational scholars taking a normative goals perspective on employee green behaviour suggest that personal norms also lead to employee green behaviour and behavioural intentions (Guiyao et al., 2023). For example, proenvironmental personal norms affect energy conservation behaviour in employees (e.g., Ruepert et al., 2016; Scherbaum et al., 2008), waste prevention and recycling (e.g., Ruepert et al., 2016), and general employee green behaviour (e.g., Mouro & Duarte, 2021). Following the NAM, proenvironmental personal norms are internalised social norms (Schwartz, 1977) and can therefore be triggered by coworkers’ proenvironmental social norms in the workplace. Research in both a private context (Kim & Seock, 2019; Klöckner, 2013; Lauper et al., 2016) and work context (Mouro & Duarte, 2021) indicates that proenvironmental social norms positively predict proenvironmental personal norms. Therefore, we developed the following hypotheses:

The Function of Organisational Environmental Strategies and the Mediating Role of Norms

Studies consulting the presence of an organisational environmental strategy as an antecedent to individual-level employee green behavioural intentions have found a positive link between them, especially indirectly via a green psychological climate (Mouro & Duarte, 2021; Norton et al., 2014, 2016; Ramus & Steger, 2000). Within the management literature, most studies are grounded in green HRM practices, such as green recruitment or training to implement environmental organisational strategies (Zacher et al., 2022), or transformational leadership to encourage and support employee green behaviour and behavioural intentions (Chen & Wu, 2022). For example, Dumont et al. (2016) found both a direct and indirect positive connection between green HRM practices and in-role employee green behaviour, but only an indirect link between green HRM and extrarole behaviour, though a psychological green climate. Thus, it seems that research from several research streams within the greening organisations or corporate greening field (e.g., organisational psychology and behaviour, green HRM and CSR) agrees that the effectiveness of organisational environmental strategies and practices on employee-level outcomes is particularly affected by psychological mechanisms such as norms (Zacher et al., 2022).

Assuming that personal norms are internalised social norms, a stream of research within environmental psychology integrates proenvironmental personal norms with proenvironmental social norms to predict behaviour and behavioural intentions (Bamberg et al., 2007; Cialdini & Jacobson, 2021). If personal norms are internalised social norms, one would expect them to mediate the impact of social norms on behaviour and behavioural intentions. A recent meta-analysis showed that proenvironmental personal norms seem to partially mediate the impact of proenvironmental social norms on proenvironmental behaviour, while social norms also have a unique (although smaller) effect on behaviour (Helferich et al., 2023). Thus, the role of proenvironmental personal norms as a mediator between proenvironmental social norms and proenvironmental behaviour is well established in the private sphere (e.g., Doran & Larsen, 2016; Kim & Seock, 2019). However, few studies have investigated the potential mediating mechanism of proenvironmental personal norms in relation to coworkers’ proenvironmental social norms and employee green behaviour and behavioural intentions in the workplace (Helferich et al., 2023; Niemiec et al., 2020). To the best of our knowledge, only one other study has looked at environmental organisational strategy and employee green behaviour in relation to proenvironmental social descriptive norms and personal norms (Mouro & Duarte, 2021). However, they focused on general proenvironmental behaviours in the workplace and did not explore the sequential effect of norms; therefore, it is still unclear how context-specific environmental strategies and high-intensity employee green behavioural intentions are connected. Thus, we predict that:

Sample and Procedure

The data were collected through an online survey at different time points from 2020 to 2022. The survey was first evaluated and revised by running a pilot with project partners. In 2020, the final full-scale survey was indirectly distributed through partner networks and other relevant channels, targeting any employee within the Norwegian construction industry who could say something about the specific topic. The survey (identical to the survey distributed in 2020) was reopened for 2 months in 2021 and 2022 due to a low response rate and the power to perform Structural Equation Modelling (SEM) analysis. Key demographic characteristics are illustrated in Table 1.

OPEN IN VIEWER

Table 1.

Demographics of Survey Respondents (N = 260).

Characteristics	Frequency (n)	Percent (%)
Profession
Municipal building owner/developer	14	5.4
Public/governmental building owner/developer	36	13.8
Private building owner/developer	23	8.8
Contractor	22	8.5
Designer, Architect	58	22.3
Manufacturer/supplier of construction products	52	20.0
Manufacturer /supplier of reused construction products	5	1.9
Other	49	18.8
Total	259	99.6
Missing	1	0.4
Experience
No experience	46	17.7
Limited experience	79	30.4
Intermediate	88	33.8
Advanced	16	6.2
Expert	10	3.8
Total	239	91.9
Missing	21	8.1
Company size
Large (≥250 employees)	168	64.6
Medium (50–249 employees)	49	18.8
Small (10–49 employees)	30	11.5
Micro (<10 employees)	12	4.6
Total	259	99.6
Missing	1	0.4
Age
20–30	27	10.4
31–40	63	24.2
41–50	57	21.9
51–60	60	23.1
Above 60	33	12.7
Total	240	923
Missing	20	7.7
Education
High school graduate	12	4.6
College graduate	33	12.7
Bachelor's degree or equivalent	60	23.1
MSc degree or equivalent	132	50.8
Doctorate degree or equivalent	3	1.2
Total	240	92.3
Missing	20	7.7
Location
Eastern Norway	59	22.7
Western Norway	13	5.0
Southern Norway	20	7.7
Central Norway (Oslo and surrounding areas)	124	47.7
Northern Norway	14	5.4
Total	230	88.5
Missing	30	11.5
Role
CEO	13	5.0
Middle manager	55	21.2
Environmentally responsible	16	6.2
Project leader	82	31.5
Employee (administration position)	14	5.4
Employee (with operational role)	44	16.9
Other	14	5.4
Total	238	91.5
Missing	22	8.5

Note. Italicised values indicate the total number of responses for each variable and system missing responses, reported as frequency (n) and percentage (%) of the total sample size (N).

In total, 340 respondents consented to participate (80 did not respond to any survey questions, leaving a final sample of 260*). The survey was available in both Norwegian (N = 308 (238*)) and English (N = 32 (22*)). Monte Carlo Simulations suggest a minimum sample size of 200 for SEM when each latent variable consists of three indicator variables; however, power analysis is generally recommended to estimate an appropriate sample size for SEM (Kyriazos, 2018). Post hoc power analysis using the Sem Power R package (Moshagen & Bader, 2023) indicated that the power to detect an effect was between 81.0% and 99.7% likely for a sample size of 260 at the 0.05 alpha level (root mean square error of approximation (RMSEA) range = .049–.069).

Measures

This article is part of research findings from a national survey about user perspectives on the reuse of construction products within the Norwegian construction industry (see Fufa et al., 2023 for access to the complete survey). The survey was developed by the authors in collaboration with a multidisciplinary research team and contained questions related to, for example, the suitability of different materials and perceived barriers for the reuse of construction products. These variables were not included in this paper, as they were related to more technical aspects and not to the conceptual model that was developed prior to making the questionnaire. Most items were newly developed by the authors, except for questions measuring general environmental personal norms (four items adapted verbatim from Schwartz (1977)) and descriptive social norms/coworkers’ green climate (four items adapted verbatim from Norton et al., 2014). Table 2 illustrates a brief overview of the measured variables linked to the research question in this article.

OPEN IN VIEWER

Table 2.

Overview of Measured Variables and Results From Confirmatory Factor Analysis.

Variable	Items	Measure	Scale	CR	AVE	α
Observed variable
Intention to reuse.		Please rate how you intend to plan for the reuse of construction products in future projects (or support strategies)?	5-point
Latent variables
Environmental organisational strategies (EOS)	Policy (EOS1)	Does your company have an ambition or goal for reuse of construction products in the company policy?	5-point	.859	.670	.805
	Projects (EOS2)	Does your company have an ambition for reuse of construction products in specific projects?	4-point
	Future (EOS3)	Does your company have a plan to set higher goals/ ambitions for reuse of construction products in near future?	5-point
Coworkers’ social norms/proenvironmental descriptive social norms (SN) – Source: Norton et al., (2014)	Pay attention (SN1)	Employees pay attention to environmental issues.	5-point	.890	.669	.888
	Behaviour (SN2)	Employees are concerned about acting in environmentally friendly ways.
	Minimise harm (SN3)	Employees try to minimise harm to the environment.
	Care about (SN4)	Employees care about the environment.
Proenvironmental personal norms (PN) – Source: Schwartz (1977)	Moral obligation (PN1)	I feel a moral obligation to protect the environment	7-point	.935	.782	.931
	I should (PN2)	I feel that I should protect the environment
	People should(PN3)	I feel it is important that people in general protect the environment
	Values/Principles (PN4)	Because of my own values/principles, I feel an obligation to behave in an environmentally friendly way.

Note. AVE = Average Variance Extracted, CR = Composite Reliability, α = Cronbach’s Alpha. All standardised factor loadings were statistically significant (p < .001).

Data Analysis Strategy

Self-report surveys pose a risk of common method variance and bias, and in line with recommendations from a recent review by Podsakoff et al. (2024), we did not use any statistical techniques but performed multiple procedural a priori remedies to reduce the potential bias effect. First, it would not make sense in this study to gather focal variables from multiple data sources because they represent individual internal states (e.g., perceptions, norms and intentions) that are difficult for others to observe; however, we randomly separated all variables throughout the survey (Podsakoff et al., 2024). Second, we attempted to reduce task difficulty (i.e., provide examples, running a pilot with representative respondents to get feedback on ambiguity) and common scale properties (i.e., different scale formats and anchor points. (Podsakoff et al., 2024). Moreover, we gathered data at different time points, and from employees with diverse backgrounds (e.g., different occupational roles, types of company and at different geographical locations).

Data analyses were conducted using the statistical package IBM SPSS Statistics 28 (IBM Corp, 2021) and IBM SPSS Amos 28 (Arbuckle, 2021). A SEM framework was chosen because it has several advantages over traditional regression models, especially when it comes to complex mediation models (Gunzler et al., 2013), enabling the exploration of interrelationships among multiple measures at once and accounting for measurement error and missing data (Morrison et al., 2017). Variables were screened prior to data analysis. There were no signs of multicollinearity (i.e., zero-order correlations were smaller than .80, see Table 3). One behavioural intention measure showed signs of exponential distribution and was therefore omitted from the analysis. Data from the 260 participants were used in the SEM analysis. There were some missing data on items, ranging from N = 60 missing for the item ‘Does your company have an ambition for reuse of construction products in specific projects’ (EOS2) to N = 5 for the item ‘Employees pay attention to environmental issues’ (SN1) (see Table 2). To deal with missing data, SEM analysis was performed using Full Information Maximum Likelihood (FIML), because this estimation technique seems to be superior to other methods when there is no systematic pattern missingness (Enders & Bandalos, 2001). Little’s test of Missing Completely At Random (MCAR) indicated that the data were missing completely at random, X 2 (129) = 150.73, p = .09 (Little, 1988). Moreover, visual inspection of the Chi-square Q-Q plot of Mahalanobis Distance may indicate a slight deviation from multivariate normality, with one potential multivariate outlier (p < .001); however, the FIML technique is arguably reasonable with regard to parameter estimation and minor deviations from normality (Yuan & Bentler, 2007).

OPEN IN VIEWER

Table 3.

Means (M), Standard Deviations (SD) and Zero-Order Correlations Between Study Measures.

Variable	M	SD	1	2	3	4
1. Intentions to reuse	3.731	1.065	1	.321**	.135*	.438**
2. Personal norms	6.206	0.881		1	.322**	.146*
3. Social norms	3.950	0.712			1	.272**
4. Environmental strategy	3.069	0.918				1

*

Correlation is significant at the .05 level.

**

Correlation is significant at the .01 level.

We followed a two-stage modelling approach to SEM (Morrison et al., 2017). First, Confirmatory Factor Analysis was undertaken to evaluate how well items represent latent constructs, enabling the latent constructs to covariate. Second, structural models were evaluated to assess interrelations between constructs.

Model fit was assessed looking at the following parameters: (a) comparative fit index (CFI) values above 0.95 (acceptable) and 0.97 (good); (b) RMSEA values less than 0.50 (good) and 0.8 (acceptable); and (c) relative chi-square index (CMIN/DF) below 2 (good) or 3 (acceptable) (Schermelleh-Engel et al., 2003). Akaike Information Criterion (AIC) was used to compare models, in which lower values indicate a better fitting model (Schermelleh-Engel et al., 2003). Joint significance test was applied to infer indirect effects, which has been shown to be a reasonable method to evaluate mediation (Fritz & MacKinnon, 2007; Hayes & Scharkow, 2013; Leth-Steensen & Gallitto, 2016; Woody, 2011).

Descriptive Statistics and Measurement Model

Table 3 shows variable M, SD and zero-order correlations between study constructs. Correlations were positive and statistically significant; however, the association between coworkers’ proenvironmental social norms and intentions to reuse, as well as proenvironmental personal norms and environmental organisational strategy were weak. The strongest zero-order association was between strategy and intention to reuse. Intentions to reuse and proenvironmental personal and coworkers’ proenvironmental social norms were high. Perceived environmental organisational strategy for reuse in the organisation was moderate.

In addition, construct validity was evaluated through assessment of the measurement model (Figure 2). The measurement model fit statistics showed an overall good fit ( X 2 = 78.516, df = 41, p < .0001, CFI = .977, CMIN/DF = 1.915, RMSEA = 0.52, AIC = 150.516). Following the suggestions by Hair et al. (2019), all constructs showed good convergent validity as assessed by composite reliability values (CR > 0.7), average variance extracted values (AVE > 0.5) (Table 2) and statistically significant standardised factor loadings above 0.5 and below 0.95 (Figure 2). Therefore, we proceeded with the full structural model to assess the hypothesised conceptual model.

OPEN IN VIEWER

Figure 2.

Measurement Model.

Structural Model and Preliminary Proposed Hypothesis Test

Overall, the hypothesised model (Model 1) showed acceptable model fit, as shown by RMSEA, CFI and CIM/DF estimates above the more rigorous thresholds (Table 4). Model 1 explains 12% of the variance in intentions to reuse or support it (Figure 3). Environmental organisational strategies for reuse are positively related to proenvironmental descriptive social norms (β = .32, p < .001) in the workplace, which in turn relate positively to proenvironmental personal norms (β = .36, p < .001) and intentions to reuse β = .33, p < .001) (Table 5). However, statistically speaking, social norms are not significantly related to intentions to reuse (β = .03, p = 7.22) when personal norms are included in the model (Table 5). A joint significance test shows that environmental organisational strategies have an indirect effect on employee intentions to reuse through a sequence of descriptive and moral norms (i.e., the relationship is serially mediated by social norms and personal norms). Thus, increased perceptions of the presence of organisational strategies for reuse positively influence employee intentions to reuse through increased positive perception of what coworkers do and subsequent feelings of moral responsibility to act environmentally friendly.

OPEN IN VIEWER

Table 4.

Model Fit Measures for Hypothesised and Alternative Models.

Model	Paths added (+)Paths removed (-)	Model type	CFI	RMSEA [90% CI]	CMIN/DF	CMIN (DF)	AIC
Model 1 (original)	EOS → SN → PN → intention SN→intention	N/A	.952	.069 [.055, .084]	2.634	134.31 (51)	212.32
Model 2(final)	(+)EOS → intention	Nested	.976	.049 [.033, .065]	1.817	90.83 (50)	170.83
Model 3	(+)EOS → intention(-)SN → intention	Non-nested	.974	.051 [.034, .066]	1.866	95.83 (51)	173.19
Model 4	(-)SN → intention	Nested	.952	.068 [.054, .083]	2.585	134.438 (52)	210.438
Model 5	(+)EOS → intention(+)SN → intention(+)EOS → PN	Nested	.976	.049 [.033, .065]	1.831	89.7931 (49)	171.701
Model 6	(+)EOS → PN(-)SN → intention	Nested	.953	.069 [.055, .083]	2.609	133.07 (51)	211.075

Note. All chi-square tests are statistically significant (p < .001). CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval; AIC = Akaike information criterion.

OPEN IN VIEWER

Figure 3.

Full Structural Equation Model of the Preliminary Proposed Conceptual Model.

OPEN IN VIEWER

Table 5.

Structural Model Results for the Hypothesised Model (Model 1).

Hypotheses	Path	Decision
H1	EOS → SN	Supported
H2	SN → PN	Supported
H3	SN → Intentions	Not supported
H4	PN → Intentions	Supported
H5	EOS → SN → PN → Intentions	Supported

The estimated hypothesised model is linked to theory; however, there might be alternative explanations and several theoretically plausible competing models. According to a recent review (Zyphur et al., 2023), it is encouraged to compare the hypothesised model with more and less constrained alternative nested and non-nested models. We evaluated five alternative models (Models 2–6, see Table 5).

Models 4 and 6 showed acceptable model fit (estimates below the more liberal thresholds), but they were not great explanations for the effect of strategy on intentions to reuse (i.e., explains less variance in intentions than expected). Models 2, 3 and 5 showed good model fit (estimates below the more rigorous thresholds), and they explain expected amounts of variance in intentions.

Model fit values served as general guidelines alongside evaluation of model suitability based on theory and research aim, model parsimony and effect size. The hypothesised model did not include a direct path between environmental organisational strategies and intentions to reuse; however, it is theoretically plausible that strategies also have a direct effect on intentions when controlling for norms. We assessed this competing hypothesis in an alternative nested model, including an added path between organisational strategies and employee intentions to reuse (Model 2, see Figure 4). Model 2 was chosen as the final model because: (1) it has decent model fit and the lowest AIC value; (2) it is less conservative than Model 1, assuming that strategy is both directly and indirectly related to intentions to reuse; (3) it is more parsimonious than Model 5 (full model); and (4) it makes sense to keep the path between social norms and intentions when model fit is similar with and without the path, because this path is well recognised in the literature.

OPEN IN VIEWER

Figure 4.

Full Structural Equation Model of the Final Model.

The final model explains 32% of the variance in intentions to reuse, compared to 12% in Model 1. In contrast to Model 1, environmental organisational strategy is positively related to intentions to reuse (β = .48, p < .001) and social norms are negatively related to intentions to reuse (β = −.15, p = .033) in Model 2. Similarly to Model 1, environmental organisational strategy is positively related to coworkers’ proenvironmental descriptive social norms (β = .32, p < .001), which are in turn positively related to proenvironmental personal norms (β = .36, p < .001), which in turn relate positively to intentions to reuse (β = .32, p < .001). Thus, increased environmental organisational strategies for reuse in the organisation seem to have a positive effect on employees’ intentions to reuse. Part of this effect may be explained through higher proenvironmental norms among coworkers, leading to increased proenvironmental personal norms. Moreover, coworkers’ proenvironmental social norms have a negative influence on intentions to reuse, unless the effect goes through proenvironmental personal norms. Since models 1 and 2 are nested, we computed a χ 2 difference test, χ d i f f 2 = χ m 1 2 - χ m 2 2 = 134.317−90.83 = 43,487, d f d i f f = 51−50 = 1, p < .001, which indicates that Model 2 better aligns with the data than Model 1.

Interpretation of Key Findings

Following expectations, we found a moderate positive relationship between environmental organisational strategy and coworkers’ proenvironmental social norms (Hypothesis 1). This is consistent with Norton et al. (2014), who found the perceived presence of a universal green policy to influence coworkers’ proenvironmental social descriptive norms in the workplace. According to our expectations, increased employee perceptions of an environmental organisational strategy for reuse in the company led to increased observation of coworkers behaving environmentally friendly at work. Moreover, consistent with expectations, we also found a medium positive impact of coworkers’ proenvironmental social norms on proenvironmental personal norms (Hypothesis 2), which again has a moderate positive influence on intentions to reuse (Hypothesis 4). This corresponds to prior research in both private (Helferich et al., 2023) and organisational (Mouro & Duarte, 2021) circumstances. Thus, if employees perceive that coworkers behave environmentally friendly at work, they are more likely to feel a moral obligation to do the same, which subsequently increases intentions to reuse.

Contrary to expectations, we did not find a positive impact of coworkers’ proenvironmental social norms (i.e., a green psychological climate) on employee intentions to reuse (Hypothesis 3): the path between the two constructs was nonsignificant in the hypothesised model and indicated a small negative effect in the final model. Since the relationship between coworkers’ proenvironmental social norms and behaviour or behavioural intentions is well established in the literature, these results should be interpreted with some precautions. The findings are contrary to previous studies that found a small positive relationship between proenvironmental descriptive social norms and behavioural intention, both generally (Helferich et al., 2023) and in an organisational context (Mouro & Duarte, 2021; Norton et al., 2014). However, Norton et al. (2014) suggest that injunctive norms predict required employee green behaviour and descriptive norms predict voluntary employee green behaviour, respectively. Although we did not explicitly ask about required/task-related or voluntary employee green behaviour, one can argue that reuse would lie closer to work tasks in a construction project, although it might require more effort and initiative from the employee than a normal work task. Another likely reason for the contrasting results is that reuse of construction products is still at an early stage in the construction industry, so a link between reuse and coworkers’ proenvironmental social norms in the workplace (i.e., what coworkers do with regard to sustainability in general) might not be well established. It is also possible that the missing or negative relationship is due to the measurement of context-specific environmental organisational strategies (i.e., reuse of construction products) and general coworkers’ proenvironmental social norms (green psychological climate). It could be that this relationship would be stronger if the social norm measure were specifically related to reuse. Moreover, as mentioned in the introduction, the construction industry is largely project-based, meaning that colleagues might not always work closely together on projects, as projects often include external stakeholders with distinct roles. Thus, it may be that coworkers’ general proenvironmental social norms are less salient in project-based industries.

In line with expectations, the effect of environmental organisational strategies on intentions to reuse was serially mediated by coworkers’ proenvironmental social norms and proenvironmental personal norms (Hypothesis 5); however, contrary to expectations, we also found indications for a direct effect between environmental organisational strategy and intentions. It seems that observed expectations from the organisation directly as well as indirectly predict intentions to reuse construction products. However, we cannot confirm whether environmental organisational strategies affect high-intensity behavioural intentions directly or depend on the internalisation of coworkers’ descriptive proenvironmental social norms. Nevertheless, it seems clear that coworkers’ proenvironmental social norms (a green psychological climate) do not have a positive effect on intentions to reuse unless the employees adapt these normative beliefs as their own. Interestingly, environmental organisational strategies had the largest effect in this study, and adding a direct path from environmental organisational strategy to high-intensity behavioural intentions increased the explained variance by a substantial amount, and models seem to align better. This finding is somewhat in line with Dumont et al. (2016), who found evidence for both a direct effect of green strategic activities on in-role green behaviour and an indirect effect through social norms (only the latter for extrarole behaviour). However, Dumont et al. (2016) looked at green HRM practices to implement environmental organisational strategies, which is slightly different to perceptions of such strategies (Dilchert et al., 2018). Conversely, our results are somewhat inconsistent with Norton et al. (2014), who suggested that social norms explain the whole effect of green policy on behaviour (i.e., norms fully mediate the relationship). This discrepancy may reflect the differences in study measures. First, they measured universal green policy and employee green behaviour, while we measured matching context-specific strategy and behaviours (i.e., related to reuse of construction products). This may explain the stronger direct association between these two concepts in our findings. Specific company policies may have a more direct impact on behaviour, but when complexity or the level of abstraction increases, the policies or certification schemes may, to a larger extent, require cultural embeddedness for successful implementation. For example, a higher level of abstraction may lead to stakeholders operating within the ‘grey area’ of greening, where there is more noise and a higher risk of diffusion of responsibility (e.g., Flagstad et al., 2022). Several researchers have adapted the classic Bronfenbrenner and Evans (2000) system model to a corporate greening context (e.g., Flagstad et al., 2021, 2022). The original article highlights the importance of the dimensions of exposure in addition to the various levels of a system that many researchers refer to. The dimensions of exposure concern the extent to which a process will lead to the development of competence or dysfunction. In addition to frequency, duration, intensity and timing, some researchers have proposed a fifth dimension of exposure: relevance – meaning what is perceived as being important (Flagstad et al., 2021). It seems clear that specific company policies may be perceived as more important if they speak to the specific work tasks of an individual employee. The concept of relevance resembles the concept of usefulness applied by Yang et al. (2020), who found that employees who perceived that their organisation had corporate green policies reported increased positive attitudes and employee green behaviour (proactive and task-related) if they perceived the policy as useful and easy to adhere to.

Second, we added personal norms to the equation. Our finding that perceived environmental strategies have a positive effect on high-intensity green behavioural intentions, but a negative effect through social norms (Hypothesis 5a) is contradictory to previous studies’ finding a positive influence of norms (e.g., Dumont et al., 2016; Norton et al., 2014). Conversely, it is consistent with the NAM and literature suggesting that personal norms mediate the relationship between social norms and behaviour (Hypothesis 5b) (Mouro & Duarte, 2021). Thus, our findings align with research suggesting that a combination of social and personal norms is stronger than the effect of social norms alone (Cialdini & Jacobson, 2021; Helferich et al., 2023; Niemiec et al., 2020).

Theoretical and Practical Implications

This paper employs a value-oriented goal-making perspective on employee green behaviour (Guiyao et al., 2023). Hence, our findings advance the understanding of integrating the theory of normative conduct (social norm theories) (Cialdini et al., 1991) with the NAM (Schwartz, 1977) to explain the potential link between perceived organisational environmental strategies and employee green behaviour. Although these theoretical perspectives are well established in explaining proenvironmental behaviour and behavioural intentions, they do not take the specific work context into account. Drawing on the principles of the theory of normative conduct in an organisational context (similar to Norton et al., 2014), this study adds to this by considering workplace-specific contextual factors like organisational environmental strategy and psychological green climate. Thus, this paper contributes to advancing our theoretical understanding of the potential link between perceived organisational environmental strategies, employee green values (i.e., proenvironmental social norms/coworkers’ climate and proenvironmental personal norms) and high-intensity green behavioural intentions. By integrating insights from environmental and organisational psychology fields to corporate sustainability research, this study responds to calls for more interdisciplinary approaches to advance our understanding of how organisations can strengthen their efforts towards circularity and sustainability (e.g., Russo et al., 2024).

One of the main aims was to examine multiple antecedents to high-intensity green behavioural intentions to offer a deeper insight into microfoundations for employee green behavioural intentions. Such a perspective is encouraged to advance theoretical development in the corporate greening field across multiple disciplines (e.g., Aguinis & Glavas, 2012; Boiral et al., 2015b; Glavas, 2016; Gond et al., 2017; Jones et al., 2017; Russo et al., 2024; Unsworth et al., 2021; Zacher et al., 2022). Specifically, we focus on the intertwined relationship between individual employee perceptions of organizational-level green strategies, team-level psychological green climate (i.e., contextual factors) and individual proenvironmental moral norms (i.e., personal factors) to predict high-intensity green behavioural intentions. By empirically examining psychological mechanisms (proenvironmental social and personal norms in a work context) that link organisational environmental strategies and high-intensity green behavioural intentions, we fill a gap in the literature that highlights the need to understand more about the underlying mechanisms that transform organisational green strategies to individual behavioural intentions. By providing insight into these microlevel processes, our findings contribute to a more comprehensive understanding of how circular environmental organisational strategies may begin to translate into organisational practices that ultimately shape the ways organisations engage with the natural environment (e.g., influence resource use in practice).

Accordingly, organisations may benefit from supplementing environmental organisational strategies for reuse (e.g., policy documents) with psychological strategies. Although employees’ perceptions of context-specific environmental organisational strategies to reuse seem to affect intentions to reuse directly, our results show that descriptive social norms (what coworkers do – a positive green psychological climate) may have a negative influence on employee intentions to reuse, unless employees feel a strong moral obligation to behave environmentally friendly. For example, if employees perceive that the organisation has prominent strategies for reuse and that coworkers do sustainable things at work in general, it might allow them to slack off, unless a moral norm to behave environmentally friendly is activated. Previous studies support the importance of activating moral norms (e.g., Wall et al., 2007).

Moreover, our findings show the possibility that environmental organisational strategies for reuse do not affect personal norms directly, but through coworkers’ proenvironmental social norms. We argue that organisations may activate proenvironmental personal norms through social norm strategies in the workplace. Specifically, it seems important to make proenvironmental social norms obvious and significant in workgroups and social settings at work. Involving employees in developing and editing environmental organisational strategies for reuse is one approach for activating proenvironmental personal norms, specifically if the employee becomes aware of the consequences of not reusing and feels a personal responsibility (by being personally involved) to reuse (Schwartz, 1977). For example, employees may feel a greater moral responsibility and later intention to reuse if the company communicate why reuse is important for circularity and why their behaviour at work is crucial for the green shift in the construction industry. Research within the HRM literature argues that such awareness can be obtained through green HRM practices, such as green training (e.g., Fawehinmi et al., 2020; Saeed et al., 2019). Thus, our study adds to the management literature by suggesting that development of behaviour-specific environmental organisational strategies for reuse is crucial to promote specific high-intensity green behavioural intentions, but that managers should be aware of coworkers’ proenvironmental social norms in the workplace. Thus, managers could focus on targeting employees’ ascription of moral responsibilities through building a psychological green climate in the company, for example, through different HRM practices. In the following, we use multilevel learning networks for climate adaptation in Norway (Hauge et al., 2019) as an example of how to increase intentions to reuse. As mentioned in the introduction, several potential learning networks exist in Norway. The management may communicate and apply the organisational strategy by enrolling employees in sustainability-focused networks. Face-to-face meetings may reinforce psychological mechanisms associated with norm activation and change. Social norms spread among group participants because humans imitate and inspire each other. Making the in-group salient may generate attitude change among participants. People who interact with an in-group in a learning network influence each other both normatively and regarding knowledge; both of these processes may be facilitated.

The findings in this study highlight the role of a combination of hard structural and soft informational strategies to reduce barriers for high-intensity behavioural intentions to reuse (Steg & Vlek, 2009). Although the study was conducted in the context of circularity in the Norwegian construction industry, the findings are situated within well-established psychological mechanisms, making them relevant for environmental organisational strategies in other sectors, arguably because organisations in general are pressed to adhere to environmental issues and cultivate circular transitions. Accordingly, in early stages of green transitions – when barriers are high – focusing on promoting clear and behaviour-specific environmental organisational strategies might be particularly effective to turn the wheels; however, it might be important to sustain these behaviours with psychological strategies (Steg & Vlek, 2009) focused on activating proenvironmental personal norms through a green psychological climate in the workplace.

Collectively, these findings highlight that embedding sustainability into everyday organisational practices depends not only on hard structural solutions such as environmental policies, but critically on how employees interpret and respond to such messages. By connecting insights from environmental and organisational psychology, this study situates individual employee behavioural intentions within a broader corporate sustainability debate. By linking psychological processes to corporate sustainability, we highlight how microlevel processes in organisations could aid in translating circular goals to everyday practices – such as the reuse of construction products – through shaping employees’ perceptions, norms and motivations to engage in high-intensity green behavioural intentions. Using the construction industry as an example, managers may help translate circular strategies – such as reuse of construction products – into everyday actions by ensuring that organisational environmental ambitions and values are clearly communicated and signalled to the employees who are ultimately responsible for embedding reuse into everyday project practices.

Limitations and Future Research Directions

The present research has limitations that need to be addressed. First, the measures are based on self-reported and repeated cross-sectional questionnaire data (the survey was distributed to an independent sample at different time points to gather a larger sample suitable for SEM analysis). We acknowledge that our research could benefit from using one data wave as a reference group; however, power was limited due to sample size. Further research might extend this by evaluating the longer-term impact.

Although the results indicate that there might be benefits to developing context-specific environmental organisational strategies that match high-intensity behavioural intentions, more research is called for to validate this relationship and investigate the potential differences between general sustainability policies and behaviour-specific policies. Moreover, it was not clear whether intention to reuse could be conceptualised as in-role or extrarole behaviour in this context; however, we argue that this might lie closer to in-role and work-task related behaviour (Ciocirlan, 2017). Therefore, we encourage further studies to clearly differentiate between these concepts, especially with regard to high-intensity green behavioural intentions that potentially have a substantial impact to aid circular transitions.

We further argue that high-intensity green behavioural intentions could be a valuable measure in the early stages of circular transformation, where specific behaviours are not widespread. For example, general intentions to reuse may be particularly relevant to consider in preliminary stages, especially since there are large barriers making reuse difficult to do in practice (e.g., Knoth et al., 2022). The emphasis on intentions in the initial stages of change could also be relevant for sustainable transformations in other industries. However, the results cannot confirm whether the measured intentions to reuse would lead to actual reuse in the future, because intentions hardly ever account for all the variance in actual behaviour (Helferich et al., 2023; Norton et al., 2016). Therefore, we cannot rule out the possibility that reuse might be something employees intend to do but find too difficult in practice. However, the reuse of construction products is not yet a widespread practice in the Norwegian construction industry, which makes measuring tangible behaviour difficult at this point in time.

Furthermore, diverse types of organisations and employees with diverse roles may have dissimilar understandings of what intentions to reuse construction products mean to them. For example, managers and those working specifically with environmental issues in their job role might have a different view on this than other employees (Francoeur et al., 2019). We did not explicitly look at these potential differences and therefore encourage future studies to investigate this further. In addition, diverse types of companies in the construction industry might understand intentions to reuse in diverse ways (e.g., consultants may promote reuse in projects, building owners through contracts and architects through design). However, perceived behavioural difficulty for reuse may be similar across types of company (Fufa et al., 2023).

In addition, although we have looked at antecedents to individual-level high-intensity green behavioural intentions at different levels of analysis, this is not truly cross-level in nature, because we conceptualised environmental organisational strategy and coworkers’ proenvironmental social norms as perceptions of these constructs (Salvador & Burciaga, 2020; Zientara & Zamojska, 2018). Although we acknowledge that a systems perspective could help explain our results and argue that our measures provide a more proximal prediction of intentions to reuse as a high-intensity green behavioural intention that may lead to actual reuse in the future (Norton et al., 2016), future studies may benefit from looking at this phenomenon from an actual multilevel perspective.