Critical Success Factors for the Widespread Adoption of Virtual Alternative Dispute Resolution (VADR) in the Construction Industry: A Structural Equation Modeling Analysis

Abstract

This study explores the increasing adoption of virtual alternative dispute resolution (VADR) in the construction industry, enhancing efficiency and accessibility in dispute resolution. VADR is crucial for streamlining processes and reducing participation barriers. The study aims to investigate the critical success factors (CSFs) influencing the adoption of VADR in the construction sector. Given the rising importance of VADR technologies, understanding the key factors driving their acceptance is crucial. The background highlights the growing reliance on innovative technologies to boost operational efficiency and decision-making processes. The data for the study were collected using a cross-sectional design with online structured survey questionnaire (N = 97) from diversified construction industries. Using Smart PLS 4, structural equation modeling (SEM) was employed to test the validity, reliability, and proposed hypotheses of the study. The results showed that cost factors had the greatest impact on VADR acceptance, followed by user competence and training, procedural adaptability, and technological infrastructure. Logistical assistance and legal frameworks also had a considerable favorable impact. However, stakeholder buy-in had no significant influence on VADR implementation. The implications indicate that economic feasibility, logistical readiness, flexible procedures, supportive legal contexts, and user skills are significant factors for successful VADR integration. Governments and organizations should deploy VADR technologies to encourage innovation and operational improvement in the construction industry.

1. Introduction

Virtual alternative dispute resolution (VADR) in the construction industry enhances dispute resolution by reducing costs, speeding up processes, and increasing accessibility. It enables the efficient handling of conflicts without the need for physical meetings. VADR is rapidly transforming industries like construction and education by improving training effectiveness and operational efficiency. The allocation of resources for VADR extends beyond monetary investments and encompasses the distribution of human resources, technological assets, and support services. Several intricately intertwined factors, including cost, logistical support, and technical infrastructure, affect the adoption rates and degree of integration of VADR technologies [1,2]. The research conducted by Exon and Lee [3] demonstrated that virtual mediation can decrease expenses related to in-person dispute resolutions, such as travel and lodging costs for mediators and disputants, thereby augmenting its perceived worth. Alessa [4] emphasized the ease of use and convenience of VADR, especially in cases involving international disputes, where participants can avoid the complications of traveling across borders. Lingasabesan and Abenayake [5] emphasized the significance of providing ongoing support and updates to technology systems in order to preserve their effectiveness in the rapidly changing technological environment.

Initial investments and ongoing maintenance costs are frequently cited as major deterrents, highlighting significant research gaps in understanding how financial barriers affect the adoption of VADR. However, in-depth examinations of the cost thresholds that drive these choices are still rare, especially in understanding how these thresholds differ between industries [1,4]. Additionally, there has been insufficient research on logistical support, despite its critical role in the effective implementation and ongoing application of VADR technologies. Requirements such as supply chain dynamics, the availability of technical support, and maintenance services are crucial but frequently overlooked factors that guarantee the operational sustainability of these technologies [6,7]. Furthermore, thorough research on procedural adaptability or an organization’s capacity to incorporate VADR into current workflows is lacking, raising concerns about the scalability and flexibility of these technologies across various organizational structures [6].

Research shows the growing role of VADR in construction, but there are still barriers to implementing VADR. Wang et al. [1] examined how VR improves construction engineering training and efficiency, laying the groundwork for VADR’s potential impact on conflict resolution. Aboulata et al. [2] investigated construction project factors for choosing alternative dispute resolution methods, but they did not address the technological integration needed for VADR implementation. Gómez-Moreno [6] discussed remote arbitration communication but did not address the specific challenges of online dispute resolution unique to the construction industry. Scherer [7] provided an analytical framework for remote hearings in international arbitration, which could inform VADR protocols but did not address sector-specific disputes. Barnett and Treleaven [8] found that while the potential of VADR has been understood, critical gaps remain, particularly in aligning virtual solutions with industry-specific requirements and integrating AI and blockchain technologies. In addition, Scherer’s [7] arbitration hearing framework showed that remote proceedings reduce the need for costly physical meetings. Alessa [4] found that AI-based decision-making could greatly impact VADR system adaptability. Gómez-Moreno [6] noted that VADR regulatory and legal frameworks are underexplored without industry-specific examples. Case studies of VADR-resolved construction disputes under these frameworks would help explain its regulatory implications. Aboulata et al. [2] suggested studying how local regulatory structures affect VADR’s construction implementation.

Regional variations exist in the impact of regulatory and legal frameworks on VADR adoption. Although specific studies have examined these influences, comprehensive comparative analyses are scarce. Depending on the regulatory landscape and industry-specific legal considerations, these frameworks may help or impede the integration of advanced technologies [8,9]. The importance of stakeholder buy-in is also underappreciated, and stakeholder perceptions significantly affect technology adoption strategies. However, studies frequently do not detail how these beliefs affect organizational decisions about implementing VADR technologies [10,11].

To address these gaps, this study aims to identify and analyze the critical success factors (CSFs) that govern the adoption of VADR technologies across different industries. By focusing on the intersection of procedural, regulatory, logistical, and economic aspects, this research seeks to elucidate how these factors influence the successful integration of VADR. Specifically, this study will (1) identify the CSFs governing VADR adoption by pinpointing the successful factors that influence the adoption of VADR technologies across different industries, (2) investigate the impact of these CSFs on VADR adoption by assessing how each identified success factor affects the adoption and integration of VADR technologies, and (3) determine the most significant CSFs by evaluating and prioritizing the importance of these factors to identify which ones most significantly impact successful VADR implementation.

By addressing these objectives, this research aims to fill existing theoretical and practical gaps, providing a comprehensive framework for understanding the dynamics of VADR adoption. This enhanced understanding will be valuable for organizations and policymakers seeking to leverage VADR technologies effectively, ensuring their optimal utilization and fostering innovation in diverse professional settings.

2. Literature Review

The effectiveness of VADR adoption in the construction industry is underscored by several aspects. Key elements such as robust technical infrastructure, including reliable internet connectivity, platform compatibility, and cybersecurity, are essential for the smooth operation of VADR systems [4,12]. Universal platform operability ensures consistent user experiences across jurisdictions [11]. Additionally, user competence and continuous training are vital for effective participation, enhancing user engagement and system usability [3,5].

Compliance with legal standards and procedural adaptability are crucial for the legitimacy and trustworthiness of VADR systems [13]. Legal frameworks must ensure the validity and enforceability of VADR outcomes to build participant confidence [5]. The standardization across jurisdictions helps maintain consistency and reliability [14]. Adapting traditional dispute resolution methods to virtual settings and addressing logistical challenges, such as time zone management and equitable access to technology, are essential for improving participant engagement and satisfaction [15,16].

Furthermore, VADR improves accessibility and inclusivity by enabling remote participation, thus overcoming geographical barriers [1]. This inclusivity is further supported by continuous user training and support [11]. Technological integration, including artificial intelligence (AI) and virtual reality (VR), enhances the effectiveness of VADR by automating dispute resolution processes and improving participant engagement through realistic simulations [2,8]. VADR also contributes to environmental sustainability by reducing the need for travel and physical documentation, thus lowering the carbon footprint of dispute resolution processes [1,11].

Given these significant benefits, understanding the factors that influence the adoption of VADR in the construction industry is crucial. This section reviews the existing literature on these factors, highlighting their significance and impact on VADR adoption. By understanding these factors, stakeholders can better facilitate the integration of VADR, leading to improved operational efficiency and decision-making processes. The following are the critical factors affecting the adoption of VADR based on existing studies.

2.1. Technical Infrastructure

VADR adoption depends on robust technical infrastructure, including reliable internet connectivity to enable real-time interaction and avoid delays [1,5]. Poor internet quality can degrade VR experiences, leading to misunderstandings [1]. Blockchain technology (BCT) can enhance internet reliability by decentralizing data management [8]. Platform compatibility, supporting multiple operating systems and devices, is crucial for VADR [6,14], Effective VADR also requires robust cybersecurity measures to protect data and privacy. AI-powered systems can enhance security by identifying and responding to threats in real time [4,11]. Holistic cybersecurity policies and practices are essential for maintaining system integrity [17]. In addition, Garon [14] emphasized the need to harmonize virtual platforms with legal frameworks to make VADR systems technologically compatible and compliant with various jurisdictions. According to Exon and Lee [3], online mediation’s effectiveness depends on platform usability and accessibility.

2.2. User Competence and Training

VADR success hinges on user competence and training, which ensure effective and confident platform use. Participants need familiarity with necessary technologies and processes to reduce anxiety and resistance, thus fostering confidence [13]. Understanding virtual environments is crucial for complex negotiations [14], and technology familiarity enhances trust in the process [12]. Effective VADR adoption requires well-designed training programs that cover platform technology and procedural knowledge [15]. Practical experience is essential, as theoretical knowledge alone is insufficient [16]. Ongoing support, including AI-driven tools for real-time assistance and responsive helpdesks, ensures that users receive the necessary help and can share best practices through forums [6,12].

Technical experience is crucial in training programs, as theoretical knowledge alone cannot prepare participants for real disputes [16]. Maintaining VADR platforms requires ongoing support including technical assistance, user education, and platform improvement. Zeleznikow [12] discussed how AI-driven tools can provide real-time support and predictive assistance to help users navigate the system and avoid common errors.

2.3. Regulatory and Legal Framework

An effective regulatory and legal framework is crucial for the widespread adoption of VADR, addressing compliance, validity and enforceability, and standardization. Compliance with relevant laws and regulations ensures VADR system integration and acceptance [5,14]. VADR systems must adhere to international standards and have verifiable and auditable frameworks [6,7]. Legal frameworks must ensure the enforceability of decisions made in virtual environments to build trust [3,18]. BCT can enhance enforceability by securely documenting decisions [8]. AI can help standardize and customize dispute resolution, ensuring consistency across different legal systems [12]. Standardizing procedures aids in transparent and robust decision-making [1,19].

2.4. Procedural Adaptability

Successful VADR implementation requires a robust procedural adaptability framework to transition traditional dispute resolution methods to virtual environments effectively [14]. This involves rethinking procedures for the virtual setting, not just transferring them online, and adapting courtroom protocols to maintain decorum and fairness [16]. Effective virtual communication is crucial for presenting arguments and evidence, with AI enhancing clarity [1,4]. Technologies like telepresence can improve mediator efficiency and participant satisfaction [3]. Blockchain technology (BCT) ensures the verification and authenticity of digital evidence [6]. Additionally, AI aids in virtual cross-examinations and document authentication, meeting legal criteria for enforceable results [12,20].

2.5. Logistical Support

Robust logistical support is essential for widespread VADR adoption, addressing challenges such as time zone management, technology access, and participant engagement. Effective global participation requires scheduling flexibility to accommodate different time zones, enhancing VADR accessibility [14,16]. BCT can help manage time zone differences by securely storing procedural actions with timestamps [8]. Equitable technology access is crucial, particularly in developing regions, necessitating targeted investments and affordable solutions [5]. Providing subsidized or loaned equipment to underprivileged parties can enhance fairness [20]. Innovations like low-bandwidth solutions increase platform accessibility [11]. Lastly, maintaining updated technology platforms ensures that they accommodate various user abilities.

2.6. Stakeholder Buy-In

Stakeholder buy-in is essential for VADR adoption, requiring support from key stakeholders, perceived value, and active promotion. Legal professionals, judiciary members, technology providers, and end-users must back VADR [14]. Industry-specific endorsements, especially from the construction sector, are crucial [5]. Educational institutions and professional development programs play a vital role in preparing future legal professionals for virtual environments [1]. Training programs and certifications in VADR standardize practices, enhance technology reliability, and increase outcome predictability [6]. Efforts to promote and advocate for VADR further support its widespread acceptance and implementation.

2.7. Cost Considerations

Implementing VADR on a large scale involves significant cost factors, including initial investments, ongoing expenses, and resource allocation. Initial investments cover technological infrastructure, software licenses, and security measures [1,8]. Strategic planning in distributing these investments ensures efficient resource use, integrating user-friendly interfaces and training modules to save costs in the long term [5]. Ongoing expenses include maintenance, updates, and administrative costs, requiring continuous support to maintain system effectiveness [3,5]. Resource allocation extends to human resources and technological assets, emphasizing training and support for mediators and legal professionals [7,12]. Investments in research and development further enhance VADR’s efficiency and outcomes [20].

In conclusion, the adoption of VADR in the construction industry is influenced by various factors, including technical infrastructure, user competence and training, regulatory and legal frameworks, procedural adaptability, logistical support, participant engagement, stakeholder buy-in, perceived value, promotion and advocacy, and cost considerations. Each of these factors plays a significant role in determining the effectiveness of VADR adoption.

Table 1. Summary of the CSFs in the adoption of VADR.

Critical Success Factors	Sub-Factors	Key Insights from the Literature
Technical Infrastructure (TI)	Internet Reliability (TI1)	Essential for effective operation and interaction [1].
	Platform Compatibility (TI2)	Need for universal platform operability [21].
	Cybersecurity Measures (TI3)	Importance of robust security to protect data and ensure privacy [4].
User Competence and Training (UCT)	Participant Familiarity (UCT1)	Familiarity enhances user engagement and system usability [3].
	Training Programs (UCT2)	Training programs necessary for skill development [6].
	Ongoing Support (UCT3)	Continuous support improves proficiency and system adaptability [12].
Regulatory and Legal Framework (RLF)	Compliance (RLF1)	Must comply with existing legal standards [5].
	Validity and Enforceability (RLF2)	Ensuring outcomes are legally recognized [18].
	Standardization (RLF3)	Crucial for maintaining consistency across jurisdictions [14].
Procedural Adaptability (PA)	Transition of Traditional Methods (PA1)	Adapting traditional methods to virtual settings [15].
	Virtual Communication (PA2)	Ensuring clear and effective communication [3].
	Cross-examination and Authentication (PA3)	Authenticating identities and evidence in a virtual environment [6].
Logistical Support (LS)	Time Zone Management (LS1)	Coordinating sessions across different time zones [16].
	Technology Access (LS2)	Ensuring equitable access to necessary technology [5].
	Participant Engagement (LS3)	Keeping participants actively involved and interested [3].
Stakeholder Buy-In (SB)	Support from Key Stakeholders (SB1)	Gaining support from legal and technical professionals [14].
	Support from Key Stakeholders (SB2)	Demonstrating cost-effectiveness and efficiency [8].
	Promotion and Advocacy (SB3)	Using success stories and strategic partnerships for promotion [16].
Cost Considerations (CCs)	Initial Investment (CC1)	Upfront costs for technology and training [1].
	Ongoing Expenses (CC2)	Recurring costs of maintenance and updates [3].
	Resource Allocation (CC3)	Efficient use of financial and human resources [20].

provides a summary of these CSFs and their key insights from the literature.

3. Research Methodology

3.1. Research Design

This study uses quantitative research methodology to identify and analyze CSFs for adopting VADR in the construction industry. Initially, a thorough literature review identified key factors which formed the basis for a pilot questionnaire. Feedback from expert interviews refined the questionnaire, which was then distributed via LinkedIn and directly to industry experts. The collected data were cleaned, removing incomplete and irrelevant responses. The cleaned data were analyzed using SEM through Smart PLS 4 to test validity, reliability, and the proposed hypotheses. Ethical considerations, including confidentiality, were maintained throughout.

3.2. Data Collection and Procedures

The data collection process was meticulously planned to ensure a representative sample of industry experts. Initially, a refined questionnaire was developed based on CSFs identified through the literature review and expert interviews. To ensure clarity and relevance, the questionnaire was pre-tested and enhanced via a pilot survey involving experienced construction dispute resolution stakeholders. A comprehensive literature review on VADR adoption CSFs informed the pilot questionnaire’s constructs. To ensure clarity and relevance, a small group of industry professionals tested it. VADR and construction experts provided content validity, question phrasing, and response scale accuracy feedback. Their feedback helped clarify and align the study’s objectives. Before release, the final version was refined and validated. The questionnaire covered VADR-related topics, including demographic information, organization type, participant role, construction industry experience, and engagement with alternate dispute resolutions (ADRs). For this pilot study, six industrial stakeholders were approached, and their opinions were solicited about the survey. Feedback and comments were received on the survey items. Later, the survey items were revised to produce a final survey. Likert-scale questions assessed the various CSFs identified in Table 1.

The questionnaire was sent to construction professionals via LinkedIn and email. LinkedIn leveraged industry networks to reach a broad and relevant audience, while email was more personal and direct. Together, these methods produced a representative sample of respondents. Professionals with VADR adoption experience were optimally involved. A convenience sampling technique targeted stakeholders from project management, consultancy, and contracting organizations, as well as ADR and VADR professionals. Data collection spanned two months, with periodic reminders sent to encourage participation. The data cleaning process involved identifying and removing incomplete or irrelevant responses. Responses with missing data in key variables and inconsistent answers were excluded from the dataset. Outliers were identified using statistical methods and discarded to maintain the integrity of the analysis. Duplicate responses were removed. After the data cleaning process, 97 valid responses remained from the 117 collected, which is considered sufficient for further SEM analysis [22]. Furthermore, it has been reported that a sample size ranging from 50 to 500 is appropriate for bootstrapping in structural equation modeling (SEM) analysis [23,24]. Therefore, the study targeted 117 respondents from the construction industry. By leveraging LinkedIn and direct email distribution, the study effectively reached a broad and relevant audience, ensuring robust and representative data on the CSFs for VADR adoption.

3.3. Data Analysis

Due to its ability to handle small sample sizes, non-normal data, and complex models, Smart PLS-SEM was chosen over CB-SEM. Hair et al. [25] recommend PLS-SEM for exploratory research models that focus on prediction rather than theory testing. PLS-SEM is also more flexible when working with real-world datasets that do not meet strict normality assumptions [26]. Memon et al. [26] also recommend Smart PLS for studies that maximize explained variance and handle formative constructs. Smart PLS also bootstraps path coefficient significance, making it ideal for exploratory hypothesis testing [27].

4. Findings of the Study

4.1. Demographics Analysis

The demographics analysis, as shown in

Table 2. Demographic analysis.

Demographics	Categories	Number	%
Organization type	Large Corporation	37	38.1%
	Medium-sized Enterprise	29	29.9%
	Small Business	31	32.0%
Participants’ role	Senior Management	20	20.6%
	Middle Management	35	36.1%
	Technical Staff	25	25.8%
	Administrative Staff	17	17.5%
Participants’ experience in construction industry	Less than 5 years	18	18.6%
	5–10 years	25	25.8%
	11–20 years	34	35.1%
	More than 20 years	20	20.5%

and Figure 1, reveals a balanced distribution across organization types, with large corporates representing 38.1%, small businesses 32%, and medium-sized enterprises 29.9%. Middle management had the highest participation rate at 36.1%, indicating their significant role in industry practice implementation and evaluation. A notable portion of participants (35.1%) reported having 11–20 years of construction industry experience, suggesting a mature and knowledgeable sample. This experienced cohort, primarily consisting of senior and middle managers, is likely to provide valuable insights into the construction industry’s dynamics and challenges. The diverse distribution across organization types and levels of experience underscores the study’s robustness and enhances its validity and applicability to the industry demographic data.

4.2. Structural Equation Modeling

There are two approaches for SEM: component-based and covariance approaches. Memon [26] stated that the component-based approach (PLS-SEM) is better suited for predictive purposes than the covariance-based approach (CB-SEM), given the fact that it is a more powerful statistical tool for assessing various parameters and reinvesting in experimental variance. The main objective of this study is to test and verify a theoretical model from a forecasting perspective. It is, furthermore, assessing the underlying relationships between various CSFs and enablers. Therefore, PLS-SEM proved valuable for this study; the model created appears in Figure 2.

To effectively utilize SEM, it is essential to develop two interconnected models: (1) the measurement model and (2) the structural model. The measurement model is crucial as it evaluates how latent constructs and their observable indicators relate to one another, ensuring that the constructs are measured accurately. A latent construct is the continuous variable, and its indicators are survey items asked for each latent construct. The structural model, on the other hand, examines the relationships between the latent constructs themselves, providing insights into the causal relationships within the theoretical framework. Both models are necessary to comprehensively evaluate the data and validate the theoretical constructs. Each of these models will be discussed in the subsequent sections.

4.2.1. Measurement Model

The measurement model assesses how latent constructs and their observed indicators relate to one another. The next subsections detail the processes and statistical tests used to evaluate the measurement model’s validity and reliability, specifically focusing on convergent and discriminant validity.

Convergent Validity

Outer Loadings: Outer loadings in reflective measurement models indicate the strength of the relationship between each indicator and its corresponding construct, reflecting the extent to which each indicator effectively measures its construct [25]. An indicator with an outer loading value of 0.4 or higher is considered acceptable [26,28]. In this study, all indicators surpassed the required threshold, demonstrating strong and reliable contributions to their respective constructs. As shown in

Table 3. Construct reliability and convergent validity tests.

Scales	Items	Outer Loadings	α	CR	AVE	Full Collinearity
Cost Considerations (CCs)	CC1	0.855	0.826	0.827	0.743	3.100
	CC2	0.896
	CC3	0.832
Logistical Support (LS)	LS1	0.704	0.720	0.739	0.510	2.417
	LS2	0.779
	LS3	0.747
Procedural Adaptability (PA)	PA1	0.745	0.718	0.715	0.569	2.272
	PA2	0.883
	PA3	0.777
Regulatory and Legal Framework (RLF)	RLF1	0.740	0.748	0.750	0.586	2.556
	RLF2	0.765
	RLF3	0.791
Stakeholder Buy-In (SB)	SB1	0.732	0.741	0.795	0.562	1.124
	SB2	0.827
	SB3	0.705
Technical Infrastructure (TI)	TI1	0.705	0.737	0.764	0.574	1.727
	TI2	0.821
	TI3	0.744
User Competence and Training (UCT)	UCT1	0.728	0.702	0.728	0.559	3.211
	UCT2	0.823
	UCT3	0.778
Overall VADR Adoption	VADRA1	0.720	0.794	0.913	0.678	3.123
	VADRA2	0.756
	VADRA3	0.718
	VADRA4	0.717
	VADRA5	0.741

, all indicators exceeded the 0.4 threshold and were therefore deemed acceptable.

Cronbach’s Alpha (α): Cronbach’s alpha coefficient assesses the internal consistency of a set of items by determining how effectively they measure a specific construct [25]. The threshold for acceptable reliability is a Cronbach’s alpha value larger than 0.7 [27]. All of the study’s constructs, as shown in Table 3, either reached or were beyond the necessary Cronbach’s alpha threshold, indicating good internal consistency. Cost considerations as well as other measurement constructs had good factor loadings, Cronbach’s alpha, and composite reliability. Therefore, all constructs were considered reliable.

Composite Reliability (CR): Composite reliability is regarded as a measure of internal consistency that is more accurate than Cronbach’s alpha, as it accounts for the varying contributions of individual items [28]. The threshold for composite reliability is also 0.7 [27]. All constructs in this study achieved composite reliability values above this threshold, as shown in Table 3, indicating high reliability and internal consistency across the scales.

Average Variance Extracted (AVE): AVE assesses the convergent validity of latent constructs, measuring the amount of variance captured by a construct relative to the variance due to measurement error [27]. An AVE value above 0.5 indicates that the construct explains more than half of the variance of its indicators [28]. As demonstrated in Table 3, all constructs in this study have AVE values exceeding the 0.5 threshold, confirming good convergent validity. These results demonstrate the internal consistency and convergence of the analytical model, with each construct adequately evaluated. Furthermore, high external loadings for the constructs suggest strong correlations and relationships between the relevant elements, reinforcing the robustness of the model.

Discriminant Validity

Fornell and Larcker [29] criterion: The criterion developed by Fornell and Larcker [29] assesses the degree of correlation that exists between the construct and its AVE square root. The AVE square root for every latent variable (i.e., construct) must be higher than the corresponding construct correlation value [29,30]. The Fornell–Larcker criterion for discriminant validity in structural equation modeling requires that each construct’s square root of the AVE be greater than its correlations with other constructs [27]. Based on the matrix, all constructs met this criterion: cost considerations (0.862) surpass their highest correlation with user competence and training (0.861), logistic support (0.714) surpasses regulatory and legal framework (0.684), and procedural adaptability (0.790) surpasses regulatory and legal framework (0.679). Regulatory and legal framework AVE (0.766) surpasses its correlations, stakeholder buy-in (0.679) surpasses its highest correlation of 0.251 with logistic support, technical infrastructure (0.758), and user competence and training (0.948) significantly surpass their correlations, and VADR adoption (0.896) surpasses its highest correlation of 0.846 with cost considerations. This shows that each construct uniquely captures indicator variance and is distinct from other constructs in the model, demonstrating robust discriminant validity.

Table 4. Correlation of latent variables and discriminant validity (Fornell–Larcker).

Constructs	CC	LS	PA	RLF	SB	TI UCT VADR Adoption
CCs	0.862
LS	0.574	0.714
PA	0.491	0.570	0.790
RLF	0.449	0.684	0.679	0.766
SB	0.242	0.251	0.193	0.082	0.679
TI	0.526	0.511	0.521	0.450	0.205	0.758
UCT	0.861	0.619	0.608	0.555	0.221	0.607 0.948
VADR Adoption	0.846	0.678	0.773	0.739	0.356	0.723 0.896 0.527

Note: CCs = cost considerations; 1 = logistic support; 3 = procedural adaptability; 4 = regulatory and legal framework; 5 = stakeholder buy-in; 6 = technical infrastructure; 7 = user competence and training; 8 = virtual alternative dispute resolution.

displays the outcomes of the Fornell and Larcker criterion [29].

Cross-loading criterion: The loading value of any indicator with its construct must be greater than the loading value of every other construct on the same row, according to the cross-loading condition [29]. The matrix showing construct–indicator cross-loadings is essential for assessing the SEM discriminant validity. Indicators that load more on their constructs than others support discriminant validity. The indicators for cost considerations (0.86, 0.90, 0.83) have the highest loadings on their own construct and the highest cross-loading at 0.82 on UCT3, confirming strong discriminant validity. Like procedural adaptability, regulatory and legal framework, stakeholder buy-in, technical infrastructure, and user competence and training, the logistic support indicators (0.60, 0.78, 0.75) have stronger loadings on their intended construct than their cross-loadings. This pattern shows that each construct uniquely captures the variance of its own indicators, distinguishing it from other constructs, which is essential for model integrity and good discriminant validity.

Table 5. Cross-loadings to test the discriminant validity of indicators.

Constructs/Indicators	CCs	LS	PA	RLF	SB	TI	UCT	VADR Adoption
CC1	0.86	0.39	0.34	0.29	0.28	0.53	0.68	0.69
CC2	0.90	0.51	0.40	0.38	0.20	0.44	0.72	0.72
CC3	0.83	0.57	0.51	0.48	0.15	0.40	0.82	0.77
LS1	0.32	0.60	0.25	0.40	0.25	0.27	0.32	0.45
LS2	0.54	0.78	0.54	0.49	0.16	0.47	0.56	0.66
LS3	0.34	0.75	0.38	0.58	0.15	0.33	0.41	0.55
PA1	(0.26)	(0.08)	(0.14)	(0.01)	(0.38)	(0.01)	(0.23)	(0.22)
PA2	0.46	0.45	0.88	0.56	0.11	0.51	0.56	0.68
PA3	0.28	0.51	0.78	0.61	0.09	0.37	0.39	0.56
RLF1	0.24	0.47	0.58	0.74	(0.01)	0.31	0.34	0.52
RLF2	0.37	0.54	0.42	0.76	0.22	0.41	0.45	0.59
RLF3	0.41	0.55	0.57	0.79	(0.03)	0.31	0.47	0.59
SB1	(0.17)	(0.08)	(0.14)	(0.11)	(0.73)	(0.13)	(0.17)	(0.25)
SB2	(0.21)	(0.29)	(0.17)	(0.04)	(0.83)	(0.19)	(0.18)	(0.31)
SB3	0.02	0.03	0.03	0.14	(0.41)	0.04	0.04	(0.01)
TI1	0.31	0.40	0.28	0.33	0.17	0.70	0.38	0.47
TI2	0.55	0.51	0.50	0.39	0.20	0.82	0.58	0.68
TI3	0.28	0.21	0.37	0.30	0.08	0.74	0.38	0.45
UCT1	0.36	0.41	0.40	0.45	(0.02)	0.55	0.63	0.54
UCT2	0.83	0.57	0.51	0.48	0.15	0.40	0.82	0.77
UCT3	0.68	0.39	0.44	0.33	0.34	0.45	0.78	0.68

Note: CCs = cost considerations; 1 = logistic support; 3 = procedural adaptability; 4 = regulatory and legal framework; 5 = stakeholder buy-in; 6 = technical infrastructure; 7 = user competence and training; 8 = virtual alternative dispute resolution.

displays the cross-loading criterion findings. The fundamental unidimensional degree of each construct is demonstrated by the loading values of the constructs being higher than the loading values of the other constructs on the same row.

4.2.2. Structural Model (Path Analysis)

The structural model, also known as path analysis, examines the relationships between the constructs in the model. The significance of these relationships is determined by evaluating the path coefficients (β-values) and p-values.

Significance of path coefficients using the β-Value: The β-value coefficient describes the strength and direction of the association between constructs [26].

Table 6. Path analysis.

Paths	β Values	p Values	VIF
CC →VADR Adoption	0.282	0.00	4.100
LS →VADR Adoption	0.149	0.00	2.417
PA → VADR Adoption	0.186	0.00	2.272
RLF → VADR Adoption	0.177	0.00	2.556
SB → VADR Adoption	0.117	0.293	1.124
TI → VADR Adoption	0.161	0.00	1.727
UCT → VADR Adoption	0.225	0.00	5.211

Note: CC: cost considerations; LS: logistic support; PA: procedural adaptability; RLF: regulatory and legal framework; SB: stakeholder buy-in; TI: technical infrastructure; UCT: user competence and training.

and Figure 3 show that the leading path coefficient is for cost considerations with a β-value of 0.282, indicating a strong influence on VADR adoption. The constructs are ranked in descending order of their β-values as follows: user competence and training (β = 0.225), procedural adaptability (β = 0.186), regulatory and legal framework (β = 0.177), Technical Infrastructure (β = 0.161), logistical support (β = 0.149), and stakeholder buy-in (SB) (β = 0.117).

Significance of path coefficients using the p-value: The statistical significance of the correlations between constructs is ascertained by the p-value [26,28]. p-values of less than 0.05 for path coefficients are regarded as statistically significant [27]. The correlations between the constructs are substantial and impactful, as evidenced by the path coefficients with p-values less than or equal to 0.05, as presented in Table 3 and Table 6. Specifically, the p-values are all 0.00, except for stakeholder buy-in which has a p-value of 0.293, indicating that its relationship with VADR adoption is not statistically significant.

Variance inflation factor (VIF): The degree of collinearity between the formative elements of the constructs in the model is ascertained by the variance inflation factor, or VIF. The VIF cut-off level is ≤ 5 [26,28]. Table 6 illustrates that every construct met the VIF threshold value of <5, suggesting that the subdomains independently contribute to the higher-order construct inside the model.

Explanatory power (R² value): The explanatory power is a measure of how well the model can explain the variation of the dependent variable [26]. The smart-PLS algorithm technique is used to calculate the R² value. VADR adoption, the study’s main dependent variable, has an adjusted R² value of 0.992. This shows that the seven independent variables explain 99.2% of the variance in VADR adoption, meaning that the seven latent variables in the model were responsible for 99.2% of the variation in VADR adoption.

5. Discussion

The adoption of VADR technologies in the construction industry is influenced by various CSFs. This study has identified and analyzed these factors using SEM to provide a comprehensive understanding of their impact on VADR adoption. The following discussion is based on the analysis results, shown in Figure 4, focusing on the significant CSFs: cost considerations, logistical support, procedural adaptability, regulatory and legal frameworks, technical infrastructure, and user competence and training.

Cost considerations emerged as the most influential factor affecting VADR adoption, with a β-value of 0.282. This indicates that financial feasibility is paramount in decision-making processes regarding VADR implementation. The findings are consistent with previous studies that highlight the importance of economic factors in technology adoption in the construction industry. Addressing cost-related concerns by providing clear ROI analyses and cost-saving benefits can enhance the adoption rates of VADR. Logistical support significantly impacts VADR adoption, reflected by a β-value of 0.149. Effective logistical frameworks, including robust supply chains, technical support, and maintenance services, are crucial for the sustainable implementation of VADR technologies. This aligns with existing research emphasizing the need for logistical readiness to ensure the smooth operation and maintenance of new technologies.

Procedural adaptability, with a β-value of 0.186, indicates that the ability of organizations to integrate VADR into existing workflows is critical for its adoption. Organizations must develop flexible procedures that can accommodate the unique requirements of VADR, ensuring that traditional dispute resolution methods are effectively transitioned to virtual platforms. The regulatory and legal frameworks also play a significant role in VADR adoption, with a β-value of 0.177. Ensuring compliance with legal standards and creating supportive regulatory environments are essential for building trust and legitimacy in VADR systems. Standardizing legal frameworks across jurisdictions can facilitate the broader acceptance and implementation of VADR.

Technical infrastructure, with a β-value of 0.161, highlights the necessity of having reliable and advanced technological systems to support VADR. This includes ensuring robust internet connectivity, cybersecurity measures, and platform compatibility. The effectiveness of VADR is heavily dependent on the technical infrastructure’s capacity to support seamless and secure virtual interactions.

User competence and training, with a β-value of 0.225, are critical for successful VADR adoption. Training programs that enhance users’ technological skills and familiarity with VADR processes are essential. Continuous support and practical training can reduce resistance and increase confidence among users, thereby facilitating the effective use of VADR. Interestingly, stakeholder buy-in did not exhibit a significant impact on VADR adoption (β-value of 0.117 and p-value of 0.293). This suggests that while stakeholder support is important, it may not be as critical as the other factors analyzed. Further research could explore the nuances of stakeholder influence and identify strategies to enhance stakeholder engagement in the adoption of VADR.

The findings are consistent with those of Aboulata et al. [2], highlighting that cost-effectiveness is crucial to the construction industry’s adoption of VADR. Logistical support boosts VADR adoption, proving that VADR technology integration requires strong logistical frameworks. In addition, Avramovic et al. [30] also stressed the importance of logistics in healthcare technology deployment. The findings are also consistent with the findings of Behfar et al. [10], who stressed adaptability in conflict resolution and team outcomes. The regulatory and legal framework, emphasizing the need for technology-friendly laws, also affects VADR adoption. Hence, Lingasabesan and Abenayake [5] explored the problems of regulatory frameworks and their potential in the construction sector’s implementation of virtual alternative dispute resolution. Compared to Exon and Lee [3], who found that stakeholder engagement is essential to creating confidence and acceptance in online dispute resolution systems, stakeholder buy-in did not significantly influence VADR adoption. Wang et al. [1] noted that robust technical assistance is essential for the successful deployment of virtual reality systems in construction engineering education. User competence and training substantially affect VADR adoption, emphasizing the need for skill development and user preparation for new technologies. This supports Barnett and Treleaven [8], who underlined the importance of training and expertise in AI and blockchain conflict resolution.

6. Conclusions

This study provides a comprehensive analysis of the CSFs influencing the adoption of VADR technologies within the construction industry, employing SEM to validate the theoretical model. The research identifies several key CSFs, including cost considerations, logistical support, procedural adaptability, regulatory and legal frameworks, technical infrastructure, and user competence and training. Among these, cost considerations and user competence and training emerged as the most significant, underscoring the necessity for detailed financial assessments and robust training programs to facilitate successful VADR implementation. This study also highlights the critical role of logistical support systems, such as reliable technical support and maintenance services, which are essential for the smooth operation and sustainability of VADR technologies. Procedural adaptability is another crucial factor, necessitating that organizations adjust their existing dispute resolution processes to seamlessly incorporate VADR. Additionally, this research emphasizes the importance of adhering to legal standards and advocating for standardized regulatory frameworks to facilitate broader VADR adoption. Investing in advanced technological infrastructure, including reliable internet connectivity and robust cybersecurity measures, is also vital for the effective functioning of VADR systems.

Despite its contributions, this study is limited by its focus on the construction industry, which may not fully capture the diverse challenges and opportunities associated with VADR technologies in other sectors. The cross-sectional design of this study also restricts insights into the long-term effectiveness and sustainability of VADR adoption. Future research should address these limitations by incorporating longitudinal studies to assess the enduring impacts of VADR technologies, exploring stakeholder engagement strategies to enhance support and buy-in, and conducting comparative analyses across different regions and sectors to identify best practices and sector-specific challenges.

Cost and user competence are the biggest CSFs affecting construction industry VADR adoption, according to this study. Successful VADR implementation depends on logistical support, procedural adaptability, regulatory frameworks, and technical infrastructure. The interaction between these CSFs emphasizes the need for organizations to balance financial investment with comprehensive training, reliable infrastructure, and flexible workflows. Standardized regulations and legal compliance boost VADR system sustainability. The framework for understanding VADR adoption in construction provides valuable insights for academic research and industry practices. The findings could help policymakers and organizations improve VADR strategies, increase adoption, and improve dispute resolution efficiency by identifying key drivers and barriers. To make VADR technologies applicable across industries, future research should examine sector-specific challenges and long-term effects.