Construction Contractors’ Carbon Emissions Reduction Intention: A Study Based on Structural Equation Model

Abstract

The high carbon emissions of the construction industry affect China’s sustainable development. Therefore, reducing the carbon emissions of the construction industry is crucial for China to achieve “carbon peak” by 2030 and “carbon neutrality” by 2060. To understand the factors that affect contractors’ willingness to reduce carbon emissions. This study is based on the Theory of Planned Behavior (TPB) and incorporates three potential influencing factors: personal norms (PN), government regulation (GR), and policy support (PS). It constructs a structural equation model (SEM) to predict the influencing factors of carbon emission reduction intentions (CERI) among construction contractors. This study analyzes the key factors and mechanisms influencing construction contractors’ CERI. The study collected 311 valid questionnaires, which are suitable for SEM research, and the results indicate that: The results show that the model proposed in the study has an explanatory rate of 69% for developers’ willingness to reduce carbon emissions. The most significant influencing factor on construction contractors’ CERI is GR, followed by PS, subjective norms (SN), and perceived behavioral control (PBC). PN and behavioral attitude (BA) towards behavior do not significantly impact CERI. SN mediate the relationship between GR and CERI, while PBC mediates the relationship between PS and CERI. The research findings can guide the government and construction contractors to improve carbon emission reduction governance mechanisms and achieve peak carbon emissions and carbon neutrality.

1. Introduction

Climate change has gradually evolved into a significant challenge that requires collective efforts from all of humanity. As a continuation of the Kyoto Protocol, the Paris Agreement has provided a unified framework for global action on climate change, with the participation of 178 parties. It was swiftly implemented after its adoption at the 2015 Paris Climate Conference, and detailed implementation guidelines were finalized at the 26th United Nations Climate Change Conference in 2021, demonstrating the determination of nations to address the increasingly severe climate change. As one of the world’s largest carbon emitters, China has committed to achieving “carbon peak” by 2030 and “carbon neutrality” by 2060. Carbon peaking: By 2030, carbon dioxide emission will no longer increase to its peak and then gradually decrease. Carbon neutrality: Before 2060, through afforestation, energy conservation, and emission reduction, offset the carbon dioxide emissions generated by oneself and achieve “zero emissions” of carbon dioxide [1]. The construction industry is vital in attaining national emission targets [2]. In 2020, the total energy consumption in the national construction sector reached 2.27 billion tons of standard coal, accounting for 45.5% of the country’s total energy consumption. In 2020, the total carbon emissions from the national construction sector were 5.08 billion tons of carbon dioxide, accounting for 50.9% of the country’s total carbon emissions [3]. Furthermore, with the continuous advancement of urbanization and improving people’s living standards, this trend is expected to increase further in the coming years [4]. Additionally, according to the analysis presented in the relevant reports published by the IPCC, the construction industry is identified as the sector with the most significant potential for emission reduction among all sectors while also being the industry with the lowest cost of emissions reduction [5].

Building life cycle carbon emissions constitute a complex process encompassing three primary stages: construction, operation, and demolition. Although the operational phase accounts for 80–90% of the energy consumption throughout the building’s life cycle [6], the potential for energy reduction in building materials and construction management should be considered [7]. In other words, the behavior of contractors during the construction process is equally significant in reducing carbon emissions. The carbon emissions associated with construction contractors encompass not only on-site construction processes but also the use and transportation of equipment, as well as on-site management [8]. You et al. [9] pointed out that a significant amount of CO2 emissions is attributed to energy consumption during building material transportation, work lighting, and the construction of foundations and walls. Furthermore, by implementing appropriate construction management strategies, contractors can reduce greenhouse gas emissions without increasing financial burdens or causing project delays [10]. Malindu [11] proposed a Construction Emission Estimation Tool (CEET) through which management can obtain a comprehensive solution in the decision-making process to minimize emissions during the construction phase of the building. That is to say, the decisions of contractors have a significant impact on carbon emissions. Therefore, starting from the goals of energy conservation, emission reduction, and environmental protection, it is essential to study the factors influencing carbon emissions from construction contractors and scientifically propose energy-saving and emission reduction measures for the construction industry to achieve carbon peak and carbon neutrality in our country.

Previous studies have primarily focused on estimating carbon emissions from contractors [9,11] and exploring specific carbon reduction measures for contractors [12,13] without examining the influencing factors of carbon emission reduction intention from the perspective of behavioral willingness in construction contractors. Therefore, this study aims to comprehensively and systematically study the influencing factors of construction contractors’ carbon emission reduction intention (CERI) and identify the key factors and mechanisms that affect construction contractors’ CERI. This study is based on the Theory of Planned Behavior (TPB) to explore construction contractors’ CERI, identify the factors influencing their CERI, establish a structural equation model (SEM), and analyze the key factors and mechanisms affecting their CERI. This study fills the gap in the factors influencing construction contractors’ CERI and expands the application scope of TPB. Additionally, this study provides managerial insights to support the improvement in carbon emission reduction governance mechanisms for both the government and construction contractors, aiming to achieve carbon peak and carbon neutrality at an earlier date.

The remaining sections of this paper are organized as follows. Section 2 provides an overview of the research status on carbon emission reduction by contractors, as well as explanations of the Theory of Planned Behavior and additional factors. Section 3 outlines the basic framework and provides explanations for behavioral attitude (BA), subjective norm (SN), perceived behavioral control (PBC), personal norm (PN), government regulation (GR), and policy support (PS). The research design is presented in Section 4, followed by the analysis of research results in Section 5, and the discussion and managerial insights provided in Section 6. Finally, Section 7 summarizes the main conclusions, highlights the theoretical and practical implications, and identifies the limitations of the research. This study mainly focuses on the following issues: (1) Identify the critical factors of construction contractor CERI. (2) Identify the impact mechanism and action mechanism of the construction contractor CERI. (3) Identify key mediating roles to explore potential logical relationships between variables that belong in the model.

2. Literature Review and Theoretical Basis

2.1. Contractors’ Carbon Emission Reduction Research

2.1.1. Carbon Emission Reduction Strategies

Given the significant role of contractors’ carbon emissions in the life cycle of carbon emissions, many scholars have conducted specific research on carbon emissions in construction. For instance, Tang et al. [10] proposed a simulation approach to evaluate the effectiveness of alternative management strategies in controlling greenhouse gas emissions. They demonstrated that an appropriate selection of management strategies could reduce greenhouse gas emissions without increasing contractor costs or delaying project schedules. In this context, enhancing contractors’ CERI holds considerable value. Therefore, more scholars have been focusing on how to achieve carbon reduction in practical contexts. For instance, Chan et al. [12] argued that using low-carbon materials during the construction phase can effectively reduce the energy consumption of a building throughout its life cycle. They also highlighted that the main obstacle construction professionals face in using low-carbon building materials is the need for more information regarding material performance. Nässén et al. [14] also researched carbon dioxide emissions during construction, considering building materials, transportation, construction activities, and machine production. Yan et al. [15] calculated the four sources of greenhouse gas emissions in construction, which include the manufacturing and transportation of building materials, energy consumption of construction equipment, energy consumption for resource handling, and disposal of construction waste. To effectively assess contractors’ carbon reduction behavior in construction projects, Wong’s [16] study adopted the European Construction Research and Development Organization’s work to evaluate Australian contractors’ carbon reduction strategies. Therefore, contractors can reduce their carbon emissions in construction projects by implementing various measures such as adopting more efficient building designs and construction methods, using energy-saving materials and equipment, and enhancing monitoring and management of building energy consumption, thereby contributing to global emission reduction targets.

2.1.2. Factors Affecting Carbon Emission Reduction

Studies on carbon reduction behavior and developers’ willingness at the management decision-making level have become relatively mature. For example, Mahmoud et al. [17] developed a multi-objective optimization model that identified 134 decision variables and provided a near-optimal Pareto frontier solution balancing low cost and sustainable development performance for building stakeholders. Lam et al. [18] classified and analyzed the factors influencing the implementation of green construction specifications, including the development of green technologies, the quality and reliability of specifications, leadership and responsibility allocation, stakeholder involvement, and the improvement in assessment benchmarks. Li et al. [19] focused on project environmental practices and defined project environmental practices from a lifecycle perspective. They studied the relationships between green design, green procurement, green construction, investment recovery, and their impacts on ecological and organizational performance.

However, these studies only investigated certain engineering content and factors affecting building carbon emissions reduction, and these influencing factors were not studied from the perspective of building implementers or contractors nor from the perspective of behavioral willingness. Therefore, this study comprehensively and systematically examines the influencing factors of contractor CERI from the contractors’ perspective, aiming to fill this unexplored gap and provide new insights into the field of carbon reduction for construction contractors. Based on the TPB, this study identifies factors influencing contractors’ CERI and analyzes key factors and mechanisms influencing contractors’ CERI through establishing SEM. The study aims to reveal the underlying mechanisms of developers’ carbon reduction decision-making in the Chinese context. Based on the TPB, this study identifies factors influencing contractors’ CERI and analyzes key factors and mechanisms influencing contractors’ CERI through establishing SEM. The study aims to reveal the underlying mechanisms of developers’ carbon reduction decision-making in the Chinese context.

2.2. The Theory of Planned Behavior (TPB)

TPB is widely applied. This theory posits that behavioral intentions (BI) and actual behavior are influenced by three constructs: attitude toward the behavior attitude (BA), subjective norms (SN), and perceived behavioral control (PBC) [20]. BA towards the behavior refers to an individual’s favorable or unfavorable evaluation or degree of assessing the behavior. If individuals believe their actions benefit the environment, they are more likely to be willing to take environmentally friendly actions. On the contrary, if individuals believe their behavior harms the environment, they are less likely to act environmentally friendly. SN refers to the influence of external social factors on individual behavior, such as social expectations, norms, and values. Individuals are more likely to adopt such behavior if a behavior aligns with social expectations and norms. PBC refers to individuals’ perception of the ease or difficulty of performing the behavior of interest. When individuals perceive that engaging in environmentally friendly actions is easy, they are likelier to take such measures.

In recent years, TPB has become one of the most influential theories for understanding, predicting, and changing various behaviors. It has gained increasing attention and application in research on the relationships between beliefs, BA, BI, and behaviors in multiple fields, such as tourism, advertising, environmental management, and project management [21,22,23]. Particularly in environmental psychology, it has been increasingly promoted as a critical theory for predicting and promoting various pro-environmental behaviors [24,25,26]. For instance, Yuan et al. [27] studied the predictive factors of project managers’ intention to reduce waste based on TPB, and the results showed that BA was the strongest predictor of project managers’ intention to reduce waste. Yang et al. [28] explored the key influencing factors of green procurement behavior among developers. They pointed out that SN, PBC, and other factors can indirectly influence green procurement behavior through BI. Therefore, this study chose TPB as the theoretical basis for the construction contractor CERI. Due to the difficulty in tracking actual behaviors [27,29], this study examines the factors influencing contractors’ CERI. As an abstract macro framework, TPB often requires contextual adaptation (including cultural and social backgrounds) to enhance its explanatory power in specific situations [20]. For instance, Li et al. [30] incorporated GR, PN, and economic feasibility into the TPB model when studying the waste reduction behavior of construction contractors, constructing a more practical framework for predicting waste reduction behavior. The results showed that surface intention significantly impacted their behavior, followed by GR and PBC. This study also studies behavior intention in environmental psychology, which is also part of TPB. This study now applies it to a new field to expand its specific application scope. Therefore, drawing on studies [30,31,32], this study also attempts an appropriate adaptation of the TPB framework to improve the understanding of the mechanisms and decision logic underlying contractors’ CERI.

2.3. The Impact of Personal Norms (PN)

PN is often considered an essential factor in contractors’ pro-environmental behavior [30]. Especially in environmental behavior, the influence of moral factors on BI should not be overlooked [33]. PN plays a significant role in determining environmental behavior [34]. Effective carbon reduction behavior can mitigate environmental damage and positively impact societal and environmental quality; thus, it can be seen as an environmental conservation behavior. Therefore, it is reasonable to consider PN influencing CERI. In Kaiser’s study [33], PN was identified as the solid antecedent of BA and BI. Botetzagias et al. [35] and Wang et al. [36] found that PN significantly impacted BI and directly predicted BI and BA. Similarly, in the field of construction contractors, Li [30] indicated that personal norms have a particularly stimulating effect on contractors’ waste reduction behavior. When contractors have strong personal norms, they adopt waste-reduction behavior to fulfill their moral obligations.

2.4. The Impact of Government Regulation (GR)

GR is often considered a predictive factor for carbon reduction intention [37]. Numerous studies have shown that government supervision and corresponding laws and regulations significantly influence contractors’ pro-environmental behavior, with the role of government being critical in developing countries [16,38,39]. Awang [40] and Khan [41] argued that government regulations are the most critical driving force for companies to implement environmental measures. Regulatory pressure from the government will encourage stakeholders to execute environmentally friendly actions. In recent years, fines and penalties for violating regulations have led to increased consideration of ecologically friendly behavior among construction companies [42]. Strict environmental regulations and high penalties for non-compliance may compel the construction industry to find new approaches that improve resource utilization [43]. Similarly, Ding [44] and Nejat [45] pointed out the vital role of government regulations and corresponding monitoring in promoting positive behavior among contractors.

2.5. The Impact of Policy Support (PS)

However, Bigerna et al. [46] argued that positive incentives (subsidies, rewards, tax exemptions, and loans) are more effective than penalties in terms of environmental behavior. Most countries have implemented various subsidies to promote green development, each with specific objectives. For instance, Australia provided production subsidies for variable renewable energy plants to achieve decarbonization in the electricity market [47], and the United States allocated significant capital grants and subsidy programs to wind energy and the grid during the financial crisis’s green stimulus plan. Subsidies issued by different government departments have diverse requirements and focuses. Environmental-related subsidies may emphasize clean production and pollution reduction, while research and development-related subsidies may focus on new technologies, processes, equipment, and materials. In the context of the construction industry, Tang et al. [32] found that government incentive measures influence the vision of construction enterprises, thereby altering their sustainable development strategies. Similarly, Alwan et al. [48] highlighted that government subsidies are effective policy tools for addressing the environmental aspects of construction, and economic subsidies serve as means to provide positive incentives.

3. Research Hypotheses

3.1. Behavioral Attitude (BA)

As one of the independent determinants of BI in the TPB framework, BA refers to the positive or negative degree an individual or organization holds toward a specific behavior [20,49]. Numerous studies have shown that individuals or organizations with a positive attitude toward a behavior are likelier to have a solid BI to engage in that behavior [50]. The study defines BA as the cognitive and willingness orientation toward CERI. BA is essential for predicting pro-environmental behavior [51,52,53]. Some studies have examined the influence of BA on pro-environmental intentions in different environmental contexts, such as low-carbon intention in travel mode choice [54], energy-saving behavior [22], recycling intention [55], and construction waste reduction behavior [56]. Darko et al. [57] even highlighted the critical role of BA in driving the adoption of green building practices. Similarly, when contractors perceive the environmental benefits of carbon-reduction construction, they become more aware of the importance of carbon-reduction behavior. They are thus more willing to engage in carbon emission reduction activities. Based on this, the following hypothesis can be proposed:

Hypothesis 1 (H1).

BA has a significant favorable influence on CERI.

3.2. Subjective Norms (SN)

According to the basic tenets of the TPB theory, SN refers to the social pressure individuals feel when deciding whether to engage in a particular behavior. This study defines SN as the impact of social pressures from stakeholders and clients on contractors’ CERI. As argued by Ajzen, individuals’ behavior is likely to be influenced when they become aware of and accept the prevailing social and cultural norms [20]. Suppose construction companies recognize the pressure from government regulations, industry standards, public media, etc., regarding their environmental management practices. In that case, they will likely develop motivations aligned with these norms, thereby stimulating environmental responsibility intentions [58]. Research has indicated that Chinese construction developers have started paying attention to the green development strategies adopted by their competitors and actively adjusting their strategic plans [59]. It can be anticipated that in the future, due to market pressures and competition, more and more enterprises will engage in green activities within the construction market [28]. Based on this, the following hypothesis can be proposed:

Hypothesis 2 (H2).

SN has a significant favorable influence on CERI.

3.3. Perceived Behavioral Control (PBC)

PBC refers to individuals’ perception of the difficulty in performing a specific behavior, including knowledge, abilities, and control factors. This study defines PBC as the influence of technical, managerial, and cost-related difficulty levels on contractors’ CERI. Klöckner [49] suggested that PBC measures the extent to which an organization has the opportunity and ability to execute a particular behavior. In environmental behavior, PBC has been studied and proven to be an essential determinant of pro-environmental behavioral intentions [60,61,62]. The more an organization believes it has the resources or capabilities to implement a given behavior, the more likely it is to intend and subsequently engage in that behavior [20]. Research by Hwang et al. [63] confirmed that the experience and technical competence of construction professionals are key factors influencing the productivity of green construction projects. Zainab [64] demonstrated that PBC could significantly impact behavioral intentions when contractors deploy new designs, technologies, material components, or construction methods within a project. The more contractors perceive their ability to participate effectively in such behavior, the more likely they are to implement it directly. Based on this, the following hypothesis can be proposed:

Hypothesis 3 (H3).

PBC has a significant favorable influence on CERI.

3.4. Personal Norms (PN)

Previous research has considered carbon emission reduction behavior as an environmental behavior driven by altruistic motives and influenced by PN [34,65]. Kaiser [33] also pointed out that the influence of moral factors on intentions or behaviors should be considered in environmental behavior. For instance, Han et al. [66] demonstrated that moral norms motivate hotel customers to reduce waste and voluntarily conserve water. In the construction industry, Liu et al. [67] highlighted that stakeholders are more likely to engage in green activities when their professional ethics level improves. Therefore, contractors will adopt a positive attitude toward high-energy consumption construction practices when they realize the wastage of resources and environmental pollution associated with them. As Li et al. [68] suggested, PN plays a critical role in the green behavior of construction site managers and has a significant favorable influence on BI. Furthermore, Li [30] investigated the relationship between PN and BA of construction contractors and found that their PN significantly influences BA. Thus, in the context of carbon emission reduction activities in construction, the PN of construction personnel facilitate their BA and CERI. Based on this, the following hypotheses can be proposed:

Hypothesis 4 (H4).

PN has a significant favorable influence on BA.

Hypothesis 5 (H5).

PN has a significant favorable influence on CERI.

Hypothesis 6 (H6).

BA mediates the relationship between PN and CERI.

3.5. Government Regulation (GR)

Increasing analyzes suggest that GR positively impacts contractors’ behaviors. In construction waste management, if the government mandates contractors to landfill construction waste and imposes penalties for illegal dumping, it will significantly reduce it [69]. Therefore, government supervision and regulatory policies are crucial in promoting carbon emission reduction measures by contractors. Additionally, Yuan et al. [27] pointed out that mandatory laws and regulations positively influence construction’ SN. As the government oversees construction carbon emission management and establishes corresponding laws and regulations, the market tends to favor policy-oriented directions. When facing market pressures, primarily influenced by competition and customer demands, contractors are more inclined to enhance their environmental capabilities and adopt green building practices to meet market needs. Zhao [70] similarly explained that adequate legislation and regulations pressure contractors to adopt pro-environmental behaviors. Hence, comprehensive government laws, regulations, strict supervision, and inspections can strengthen contractors’ SN and CERI. This study proposes the following hypotheses:

Hypothesis 7 (H7).

GR has a significant favorable influence on SN.

Hypothesis 8 (H8).

GR has a significant favorable influence on CERI.

Hypothesis 9 (H9).

SN mediates the relationship between GR and CERI.

3.6. Policy Support (PS)

The government provides incentives or plans to serve as a driving force for construction companies to implement green building projects [71]. These incentive policies or plans can enhance project owners’ financial benefits through tax deductions, preferential loans, special funds, grants, rebates, and subsidies [72,73]. Financial incentives compensate for the additional costs of green building measures [74]. Government incentive policies can also be non-financial, relaxing and reducing administrative and technical requirements involved in green building project development, thereby saving time and money [75]. Other forms may include technical assistance [76], ceremonial awards, or recognition [72]. In the construction industry, the government can incentivize contractors to focus on activities that protect the environment, such as reducing construction waste and carbon emissions, through implementing economic subsidies. Reasonable subsidy amounts would encourage contractors to adopt environmentally friendly practices [67]. Furthermore, the government is committed to improving the research and development capabilities of enterprises dedicated to carbon reduction [77] and promoting activities such as remanufacturing [78]. Green technological innovations have positive environmental impacts and benefits for enterprises that do not bear the costs [79], and government support for green technology contributes to the accumulation of knowledge quality and technical capabilities in this field [80]. Therefore, based on this, the following hypotheses can be proposed for this study:

Hypothesis 10 (H10).

PS has a significant positive impact on PBC.

Hypothesis 11 (H11).

PS has a significant positive impact on CERI.

Hypothesis 12 (H12).

PBC mediates the relationship between PS and CERI.

Research on the role of PN, GR, and PS in promoting contractors’ CERI has yet to be conducted. Due to significant institutional differences between countries and a need for more explicit and consistent understanding, this issue must be examined through empirical investigations of each country’s specific circumstances. The present study utilizes the Theory of Planned Behavior (TPB) as the fundamental model framework to fill this research gap. It integrates PN, GR, and PS into the TPB model. A theoretical model is constructed to explore the key factors influencing contractors’ CERI in the construction industry. The theoretical model is illustrated in the following diagram. (See Figure 1)

4. Research Design

4.1. Questionnaire Design

This study primarily collected data through the distribution of online questionnaires. The research ensured the scientific rigor and applicability of the questionnaire; each measurement item was primarily designed using a Likert 5-point scale. Before the official survey, the research team conducted a pilot test of the scale, collected 41 pilot test data for analysis, and removed one item (BA1) with inadequate reliability and validity based on the analysis results. The remaining three items were re-sequenced from BA1 to BA3 to form the final survey questionnaire. The questionnaire consists of two main parts: (1) a survey of respondents’ background information, including work experience, education level, job position, company qualifications, and company type, and (2) a survey of seven dimensions: BA, SN, PBC, PN, GR, PS, and CERI. The measurement variables are presented in

Table 1. Survey Scale for Contractor’s Carbon Emission Reduction Intentions.

Variables	Item	Measurement Scales	Sources
CERI	CERI1	The company is willing to establish a management system to reduce carbon dioxide emissions.	[30,55,67,68]
	CERI2	The company is willing to purchase raw materials with low carbon dioxide emissions.
	CERI3	The company is willing to research and develop new technologies for reducing carbon dioxide emissions.
	CERI4	The company is willing to learn and promote low-carbon and environmental protection knowledge.
BA	BA1	The practice of green carbon emission reduction in construction contributes to improving the natural environment around the company.	[56,81]
	BA2	The company is interested in carbon emission reduction.
	BA3	The company supports carbon emission reduction in construction.
SN	SN1	Due to the specialization in low-carbon practices, the company faces pressure to adopt low-carbon construction.	[40,67,81]
	SN2	The company faces pressure for carbon management due to low-carbon standards.
	SN3	Competitors who adopt low-carbon construction methods exert pressure on the company to adopt low-carbon construction.
	SN4	The demands of clients for low-carbon practices will prompt the company to adopt carbon emission reduction in construction.
PBC	PBC1	The company is currently in good financial standing and has sufficient funds for carbon emission reduction in construction.	[30,56,67]
	PBC2	The company has professional personnel and mature technologies, enabling it to carry out carbon emission reduction in construction efficiently.
	PBC3	The company possesses rich experience in carbon emission reduction in construction and excels in carbon management.
PN	PN1	I would feel guilty if there is no implementation of low-carbon management in the construction.	[30,68]
	PN2	Adopting carbon emission reduction in construction is my moral obligation.
	PN3	I will strive to choose green and low-carbon technologies for construction.
	PN4	I will make an effort to choose green and low-carbon materials for construction.
GR	GR1	The government establishes environmental regulations for construction operations, compelling the company to adopt carbon emission reduction in construction.	[30,81,82]
	GR2	The government requires the company to improve its environmental performance.
	GR3	The government will impose penalties on environmental violations.
	GR4	The government has a well-established regulatory system for managing carbon emission reduction in construction.
PS	PS1	The government provides financial support for implementing carbon emission reduction and environmental protection measures.	[32,40,56]
	PS2	The government provides technical assistance for implementing carbon emission reduction and environmental protection measures.
	PS3	The government assists in training skills for carbon emission reduction in construction.

.

4.2. Data Collection and Sample Characteristics

This study conducted an online questionnaire survey from September to November 2022. All respondents to this study were willing to participate in this experiment. The questionnaire was communicated through WeChat to those who met the requirements of this study. All respondents answered anonymously; their personal information is confidential and does not involve sensitive issues. The questionnaire was distributed using snowball sampling. Snowball sampling is one of the most popular methods in qualitative research, with its core feature being networking and referral [83]. Fewer specific populations meet the research criteria, so snowball sampling suits this study. This study began with 40 initial contacts (seeds) who met the research criteria and were invited to participate. Then, with the consent of the participants, they were asked to recommend other contacts who met the research criteria, were also willing participants, and so on. A total of 400 questionnaires were distributed, and 364 were collected. This study found through the pre-experiment that the respondents completed their responses within at least 90 s. Therefore, the decision made in this study to complete the responses within 90 s is not considered answering seriously and is considered an invalid questionnaire. After excluding 26 incomplete and 53 questionnaires with a response time of fewer than 90 s, the final valid sample size was 311, resulting in an effective response rate of 85.44%.

Regarding the sample size of the questionnaire, Barrett suggested that the sample size should be larger than 200. However, the study also pointed out that when SEM is performed using the built-in maximum likelihood method, the chi-square value inflates significantly when the sample size exceeds 500, leading to poor model fit [84]. Since the present study primarily employed the maximum likelihood method for parameter estimation, the sample size should be greater than 200 but less than 500.

Descriptive statistical analysis was performed on the sample’s demographic characteristics, and the results are presented in

Table 2. Distribution of Sample Characteristics.

Variable	Option	Frequency	Percent
Years of work experience	5 years and below	83	26.7%
	5–10 years	106	34.1%
	10–20 years	83	26.7%
	20 years and more	39	12.5%
Educational level	Senior middle school and below	22	7.1%
	Junior college	93	29.9%
	Undergraduate	166	53.4%
	Graduate and above	30	9.6%
Corporate position	Construction workers	121	38.9%
	Site management workers	130	41.8%
	Project managers	30	9.6%
	Others	30	9.6%
Company qualification	First-class qualification	153	49.2%
	Second-class qualification	116	37.3%
	Third-class qualification	42	13.5%
Type of Enterprise	State-owned enterprise	146	46.9%
	Private enterprise	104	33.4%
	The foreign-invested or joint venture	45	14.5%
	Other	16	5.1%

.

5. Result

5.1. Analysis of Reliability and Validity

To ensure the internal consistency and validity of the scale data, this study analyzed the reliability and validity of the data used in model construction. This study primarily analyzed the items’ reliability and validity through convergent and discriminant validity. The analysis was performed using SPSS 23.0 software, with a KMO value of 0.927 > 0.8 and Bartlett’s sphericity test chi-square value of 4024.739, p < 0.01, indicating that the data passed the significance test and were suitable for factor analysis and validity testing [85].

Convergent validity reflects the model’s internal consistency, as indicated by composite reliability (CR) and average variance extracted (AVE). When CR is greater than 0.7, AVE is greater than 0.5 and Variance Inflation Factor (VIF) less than 3, it indicates good internal consistency of the model [86,87]. results shown in

Table 3. Summary of reliability and convergent validity analysis.

Potential Variables	Observed Variable	Unstd.	S.E.	Z-Values	p	Std.	SMC	CR	AVE	VIF	Cronbach’s α Coefficient
GR	GR1	1				0.736	0.542	0.863	0.612	1.342	0.862
	GR2	1.124	0.078	14.441	***	0.861	0.741
	GR3	1.061	0.078	13.663	***	0.807	0.651
	GR4	0.926	0.076	12.145	***	0.718	0.516
PBC	PBC3	1				0.739	0.546	0.787	0.552	1.622	0.787
	PBC2	1.067	0.095	11.273	***	0.770	0.593
	PBC1	1.027	0.095	10.859	***	0.720	0.518
CERI	CERI1	1				0.764	0.584	0.839	0.566	Dependent variable	0.882
	CERI2	1.028	0.083	12.38	***	0.718	0.516
	CERI3	1.063	0.081	13.045	***	0.754	0.569
	CERI4	1.062	0.079	13.363	***	0.772	0.596
SN	SN1	1				0.730	0.533	0.854	0.595	1.729	0.853
	SN2	1.052	0.082	12.799	***	0.789	0.623
	SN3	1.067	0.082	12.994	***	0.803	0.645
	SN4	1.009	0.081	12.391	***	0.761	0.579
PN	PN4	1				0.832	0.692	0.845	0.578	1.635	0.844
	PN3	0.953	0.068	14.081	***	0.772	0.596
	PN2	0.836	0.063	13.216	***	0.729	0.531
	PN1	0.826	0.065	12.654	***	0.702	0.493
BA	BA3	1				0.774	0.599	0.761	0.518	1.428	0.755
	BA2	0.792	0.084	9.432	***	0.616	0.379
	BA1	1.005	0.093	10.75	***	0.758	0.575
PS	PS3	1				0.748	0.560	0.810	0.586	1.579	0.809
	PS2	0.989	0.082	12.032	***	0.772	0.596
	PS1	1.004	0.083	12.081	***	0.777	0.604

Note: *** indicates p < 0.001.

by conducting confirmatory factor analysis using AMOS. It can be observed that CR values for all dimensions are greater than 0.7, and AVE values are greater than 0.5, indicating good internal consistency of the model. Discriminant validity reflects the distinctiveness among latent variables. The analysis results in

Table 4. Discriminant Validity Table.

	BA	PN	SN	CERI	PBC	GR	PS
BA	0.720
PN	0.608	0.760
SN	0.486	0.493	0.771
CERI	0.443	0.554	0.699	0.752
PBC	0.365	0.463	0.649	0.689	0.743
GR	0.183	0.388	0.505	0.74	0.497	0.783
PS	0.402	0.555	0.575	0.679	0.621	0.395	0.766

Note: The diagonal number is the root value of the factor AVE.

show that the correlation coefficients between any two variables are smaller than the square roots of their respective AVEs, indicating good discriminant validity of the scale. The research model in this study demonstrates good discriminant validity.

5.2. Model Fitting

This study constructed a structural equation model (SEM) for hypothesis testing. After incorporating the observed variables, the resulting model structure is shown in the diagram below. The structural equation model consists of seven latent variables: CERI, BA, SN, PBC, PN, GR, and PS.

Chi-square tests the discrepancy between the sample and the matrices of covariance fitted in the model. CMIN/df value of 3 or less is considered an excellent model fit measure [88]. Goodness-of-fit statistic (GFI) estimates the proportion of the variance provided by the projected covariance of the population. It ranges from 0 to 1. In general, the recommended threshold is 0.80 [89]. Adjusted Goodness-of-fit statistic (AGFI) tries to adjust the GFI with degrees of freedom. It also ranges from 0 to 1. Generally, the widely recommended threshold is 0.80 [88]. The Comparative Fit Index (CFI) compares the model fit with a null or independent model. The major difference is that it talks about latent factors rather than indicators. A threshold value of 0.90 and above suggests a good model fit [88]. Root Mean Square Error of Approximation (RMSEA) is considered the best informative fit index. It goes for an optimal number of parameters (lesser) to fit the final population covariance matrix. An excellent model fit should have an RMSEA value of 0.08 or less [90]. The Normed Fit Index (NFI) index evaluated the model by comparing the chi-square value of the model and the same null or independent model. A threshold value of 0.80 and above suggests a good model fit [90]. To determine the best-fitting structural equation model, this study primarily relied on the indicators and fit criteria in

Table 5. Discriminant Validity Table.

Index	Model Value	Reference Standard	Conclusion	Source
CMIN/DF	2.017	1–3 excellent, 3–5 good	excellent	[88]
GFI	0.881	>0.9 excellent, >0.8 good	good	[89]
AGFI	0.855	>0.9 excellent, >0.8 good	excellent	[88]
CFI	0.93	>0.9 excellent, >0.8 good	excellent	[88]
RMSEA	0.057	<0.08 excellent, <0.1 good	excellent	[90]
NFI	0.871	>0.9 excellent, >0.8 good	good	[90]

to evaluate the model fit. Comparative analysis revealed that the goodness-of-fit indices met the reference standards, indicating a good model fit and sufficient adaptability to the collected data.

The specific model is shown in Figure 2.

5.3. Model Path Analysis

The results of the hypothesis testing are presented in

Table 6. Path Analysis of the Structural Equation Model.

Hypothesis Path	Dependent Variable	Independent Variable	Estimate	S.E.	CR	p	R2	Results
H1	SN	GR	0.505	0.07	7.203	***	0.258	Support
H2	BA	PN	0.518	0.061	8.481	***	0.369	Support
H3	PBC	PS	0.547	0.068	7.993	***	0.388	Support
H4	CERI	PBC	0.169	0.063	2.698	**	0.695	Support
H5		GR	0.418	0.052	8.018	***		Support
H6		SN	0.171	0.045	3.779	***		Support
H7		PS	0.231	0.056	4.143	***		Support
H8		BA	0.103	0.055	1.868	0.062		Not supported
H9		PN	0.054	0.045	1.196	0.232		Not supported

Note: ** indicates p < 0.01, *** indicates p < 0.001.

, indicating that only H8 and H9 hypotheses were rejected, while the remaining hypotheses were supported. This study will discuss and analyze the significant paths. From the analysis of Table 6, it can be observed that out of the nine hypotheses investigated in this study, seven were supported. Specifically, GR (β = 0.418, p < 0.001), PS (β = 0.231, p < 0.001), and SN (β = 0.171, p < 0.001) significantly influenced CERI. GR (β = 0.505, p < 0.001) significantly influenced SN, PN (β = 0.518, p < 0.001) significantly influenced BA, PS (β = 0.505, p < 0.001) significantly influenced PBC, and PBC (β = 0.505, p < 0.01) influenced CERI. It is necessary to explain the meaning of path coefficients. Path coefficients (β) visually indicate the strength of the relationships between variables. Taking GR’s influence on CERI as an example (β = 0.418), for every one-unit increase in GR, CERI is expected to increase by 0.404 units. Therefore, the study supports the original hypotheses H1, H2, H3, H4, H5, H6, and H7, while H8 and H9 were not supported in this research.

5.4. Bootstrap Mediation Analysis

In order to further investigate the relationships among variables, this study conducted mediation analysis on four variables: GR, CERI, SN, and PS. The results of the analysis are presented in

Table 7. Results of Bootstarp Mediation Analysis.

Path Analysis	Effect	Estimate	Lower	Upper	p	Percentage	Results
GR-CERI	Indirect effect	0.087	0.046	0.143	***	17%	Partial Mediation
	Direct effect	0.418	0.334	0.519	***	83%
	Total effect	0.504	0.408	0.612	***
PN-CERI	Indirect effect	0.053	0.003	0.111	0.079	50%	No Mediation Effect
	Direct effect	0.054	−0.039	0.144	0.358	50%
	Total effect	0.107	0.043	0.178	**
PS-CERI	Indirect effect	0.092	0.039	0.16	**	28%	Partial Mediation
	Direct effect	0.231	0.131	0.357	***	71%
	Total effect	0.324	0.229	0.442	***

Note: ** indicates p < 0.01, *** indicates p < 0.001.

. It can be observed that CR significantly mediates the path between SN and CERI (p < 0.001), and PS significantly mediates the path between BA and CERI.

5.5. Robustness Test of the Model

In this study, Years of work experience, corporate position and type of enterprise were introduced into the model as control variables to the test robustness of the hypothesis model. The test results areshown in Figure 3.

Figure 3 shows that although control variables such as years of work experience, corporate position and type of enterprise were introduced, the relationship and significance level of each factor in the model are consistent with the conclusion of the hypothesis test results mentioned above. Meanwhile, the test results of the impact of each control variable on the actual use of facial recognition were insignificant, indicating that the model passed the robustness test.

6. Discussion

6.1. Factors Influencing CERI

6.1.1. The Impact of GR on CERI

From Table 6, it can be observed that GR has the greatest impact on CERI. It has a significant direct effect on CERI and an indirect effect through SN. Table 7 shows that the direct effect is 0.418, and the total effect is 0.504. The construction industry in China is more sensitive to government regulations than other industries [91], leading to a strong correlation between GR and contractors’ CERI. Consistent with the findings of this study, many other studies have also indicated that government laws and regulations are key factors influencing contractors’ green and environmentally friendly behaviors [30,38]. The study [92] results demonstrate that mandatory government institutional arrangements can stimulate companies’ willingness to engage in low-carbon production.

The relevant government departments should gradually specify the low-carbon requirements contractors should meet during construction through policies, legal norms, standards, and other documents, based on assessing the average level of low-carbon construction among contractors in China. The government environmental regulatory agencies can supervise the low-carbon environmental behaviors of construction-related enterprises directly or by commissioning third parties, ensuring the effective implementation of low-carbon environmental design schemes during construction and the effective implementation of documents related to contractors and project green environmental aspects.

6.1.2. The Impact of PS on CERI

Moreover, PS has a significant impact on CERI as well. Table 7 reveals a direct effect of 0.231 and a total effect of 0.324. Consistent with the findings of Li et al. [93], government support policies for environmental behaviors not only contribute to enhancing the environmental governance capacity of enterprises but also guide investments and the adoption of innovative technologies. Implementing incentive policies can enhance the capacity and motivation of enterprises to innovate waste-reduction technologies, reduce the demand for raw materials, increase the utilization of recycled resources, and simultaneously decrease environmental pollution.

The incentive mechanisms established by the government through relevant policies are the driving force behind the development of low-carbon construction [71]. The government’s favorable measures significantly promote the reform of relevant enterprises towards low-carbon construction. Policies and regulations introduced by government departments at all levels should focus on transforming construction entities and strongly support low-carbon construction enterprises. To promote the industrialization of low-carbon construction, it is crucial to provide police protection and urge relevant authorities to prioritize approving construction progress-restricted documents, dedicated funds, bank loans, and infrastructure. Additionally, establishing corresponding incentives for construction entities with advanced technologies is essential.

6.1.3. The Impact of SN on CERI

SN has an effect of 0.171 on CERI. Similar to the results in [29], he indicated that SN positively impacted the contractor’s willingness to recycle construction waste. This once again validates the TPB proposed by Ajzen [20]. The driving forces behind construction contractors’ implementation of carbon emission reduction management in projects are the requirements of clients, competition among peers, and normative pressures from the public. As one of the primary external drivers for low-carbon management in construction projects, market competition should play a significant role. However, this external force’s impact is weakening due to the need for standardized market competition in our country’s construction industry. Therefore, to enhance the driving force behind carbon emission reduction management in construction projects, it is crucial to establish a fair market order and create an environment of survival of the fittest and free competition. Additionally, it is vital to strengthen the low-carbon awareness of clients and leverage their role in guiding and supervising. The intense market competition requires all types of construction companies to prioritize the needs of project clients, and the low-carbon demands of clients can drive construction companies to implement carbon emission reduction management. Therefore, strengthening the low-carbon awareness of project clients, making them aware of the benefits of implementing low-carbon management and construction, and leveraging their role in guiding and coordinating efforts, can promote implementing carbon emission reduction management in construction companies.

6.1.4. The Impact of PBC on CERI

Data analysis confirms that PBC is another positive factor influencing CERI. Li [68] also demonstrated that PBC is essential in reducing waste for construction contractor employees. What is different is that in this study, its positive effect was minimal (the effect size was only 0.169). It may be because the effects of GR and PS weaken the impact of PBC on de CERI. Construction contractors with sufficient technical expertise, management capabilities, and financial resources can only engage in carbon emission reduction activities. The flexible application of technology can influence contractors’ willingness to adopt low-carbon practices and help mitigate construction difficulties. Therefore, the development and application of technology are crucial in promoting carbon emission reduction practices across all stages of construction. The advancement and implementation of low-carbon building practices depend on technological innovation and application, emphasizing the need to concentrate on technological innovation. With a focus on technology, it is essential to enhance the innovation awareness of all parties involved, recognize technological gaps, and develop technologies suitable for China’s specific characteristics. In addition to technological factors, low-carbon construction can be achieved through innovations in construction plans, construction concepts, and management methods.

6.2. The Impact of Mediation

GR significantly influences CERI through its significant impact on SN. The mediation accounts for 83% of the total effect. The government is considered the primary source of pressure for businesses, with various manifestations. Companies must meet government expectations to reduce uncertainty and achieve their goals [94]. The healthy and orderly development of the market also relies on government regulations and guidance. In the market environment of construction, the government should guide the market to protect the environment and reduce carbon emissions based on fair competition under the socialist market economy system. The government can also use the relationships between stakeholders to form constraints among various units in the low-carbon aspect of the project, such as requiring the design unit to design low-carbon plans, the construction unit to execute according to the design plan, and the public to supervise the construction unit to execute.

PS significantly influences CERI through its significant impact on PBC, with a mediation effect of 72%. It indicates that the government supports construction contractors in implementing low-carbon project management through tax incentives, loan guarantees, and financial subsidies [32,48]. Implementing low-carbon management in construction projects requires exploring new management models and applying new management techniques, tools, and methods in the project management process. It represents a transformation and challenge to the traditional project management approach. However, this transformation may increase various costs and risks for enterprises in the initial stage. Therefore, for business managers, if the existing management and operational models can achieve the desired objectives, they may be unwilling to make any changes, even if such changes potentially improve productivity and management capabilities. During the early stages of implementing low-carbon management in projects, the government can employ restorative measures such as tax incentives, loan guarantees, and financial subsidies to minimize the additional costs and risks construction companies face due to implementing low-carbon management. Moreover, companies that achieve better results in implementing low-carbon management should receive more incredible policy support to encourage more construction companies to apply low-carbon thinking in their practical project management. Furthermore, providing substantial financial subsidies and tax incentives implies contractors can save project costs and minimize penalties. Therefore, PS is an essential environmental policy that promotes contractors’ CERI and significantly strengthens their PBC capabilities, enabling them to engage in carbon reduction activities more effectively.

Although BA is considered an important predictor of pro-environmental behavior, this study found no significant impact of BA on CERI. Similarly, the hypothesis that PN directly impacts CERI was also rejected. This suggests that the influence of internal moral norms on CERI is less significant than expected. It contradicts the findings of [95,96], where BA and PN have been identified as essential factors in low-carbon lifestyle and travel. It can be argued that this is because low-carbon social behavior is an individual act that individuals can control. In contrast, carbon reduction in construction is a collective behavior driven by shared interests. It is easy to understand. China is still a developing country where many people strive to improve their quality of life, especially in the construction industry, where wages are generally low. According to Maslow’s hierarchy of needs, spiritual needs are only emphasized when basic needs are met.

7. Conclusions and Limitations

7.1. Conclusions

This study synthesized the factors influencing contractors’ CERI through a literature review and constructed a conceptual model and structural equation model for the factors based on existing research. Then, a questionnaire survey was conducted to collect data, which were processed using SPSS and AMOS software to verify the hypotheses in the model and analyze the mechanisms of contractors’ CERI. Based on the results of the model analysis, this paper draws the following conclusions:

• The results show that GR has the most significant impact on CERI. The relevant government departments not only explicitly require low-carbon requirements through policies, legal norms, standards and other documents, but more importantly, they carry out strict supervision, punish construction contractors who do not meet low-carbon standards, and suspend production or even suspend business licenses for enterprises that seriously pollute the environment. Only by strictly implementing policies can relevant government departments contribute to achieving the goals of “carbon peaking” and “carbon neutrality”.
• PS, SN, PBC have a significant impact on CERI. In addition to issuing strict regulations and supervision, economic measures can be taken to enhance contractors’ awareness and autonomy in carbon reduction, such as tax incentives, financial subsidies and green construction certification. Contractors can learn advanced construction technologies from each other through organizing enterprise forums, international exchanges and cooperation, and improve their low-carbon awareness through communication. In addition to technical factors, contractors can also achieve innovation by reducing construction plans, construction concepts, management methods, and other aspects similar to recyclable waste.
• The government indirectly affects CERI by influencing SN and PBC. The government can entrust third parties and stakeholders to supervise contractors. The government’s incentive policies can also improve the ability and power of enterprises to innovate and reduce waste technology, increase the use of renewable resources, and reduce environmental pollution.

7.2. Theoretical Implications

TPB is an essential framework of social sciences for predicting behavioral intentions, and its value has been widely recognized in the environmental context [30,67,85]. However, this study’s significant theoretical contribution lies in improving this theory by logically incorporating GR, PS, and PN variables into the research model. Furthermore, the improved theoretical framework was applied to explain the factors influencing contractors’ CERI. This study’s most significant theoretical value lies in extending the specific application contexts and boundaries of the theory, thereby expanding its applicability scope and confirming TPB’s flexibility and adaptability.

Furthermore, the theoretical significance of this study lies in deepening the understanding of TPB through in-depth research and providing new insights and methods for predicting and explaining CERI in the field of construction contractors. Additionally, this study provides valuable references and guidance for construction contractors’ research and application of carbon reduction. Specifically, it offers a theoretical framework and research model that can be used for future studies, further promoting the development of construction contractors in the field of carbon reduction and providing insights for achieving carbon neutrality at an earlier stage.

7.3. Practical Implications

This study also makes significant contributions at the practical level. Firstly, the research findings provide strong evidence for the government and construction companies. This study investigates the influencing factors of contractors’ CERI, providing relevant strategies for the government and construction companies to pursue carbon reduction.

Furthermore, this study considers GR and PS as two potential influencing factors. It reveals that they directly impact contractors’ CERI and have a significant indirect influence through SN and PBC. This enriches the research on the influencing factors of urban green space conservation and contractors’ CERI, demonstrating the government’s decisive role in future environmental decision-making. Overall, this study provides new insights into contractors’ CERI, laying a new theoretical foundation for the study of contractors’ carbon emission reduction behavior and sustainable development.

7.4. Limitations and Suggestions for Further Studies

This study has several limitations. Firstly, the empirical analysis is based on data from construction companies in Wuhan rather than including a range of other cities in China. While most cities in China have similar policies and operate under the central government’s guidance, the social environment for carbon emissions reduction can vary across different regions and cities. Therefore, the research findings may need adjustments before being applied to other cities in China. Secondly, this study mainly focuses on capturing the construction’ CERI based on TPB without directly examining their actual behavior. Further research could explore the carbon reduction behaviors of contractors to gain a more comprehensive understanding of the factors influencing carbon reduction behavior. Finally, regarding the conceptual framework, this study only considers GR, PS, and PN based on TPB, potentially overlooking the influence of other factors, such as economic benefits, organizational culture, waste reduction and recyclability factors. Future research should consider a broader range of potential influencing factors to enrich the study and broaden the research scope. Despite these limitations, this study provides a guideline for future research.