How to Foster Project Organization Resilience in the Construction Industry: The Role of Data Governance Capabilities

Abstract

With the ongoing changes in global economic and social environments, project organization resilience has emerged as a core competency in addressing environmental uncertainty, thereby becoming a prominent research area within construction management. This resilience is intricately linked to data resources; however, construction enterprises face systemic challenges in governance—fragmented data standards, siloed storage, and underutilized analytics—which limit their ability to translate data into crisis-responsive actions. Consequently, it is of paramount importance to investigate how data governance capabilities influence project organization resilience. This study initially identifies five dimensions of data governance capabilities within construction enterprises, spanning the three stages of planning, flow, and application: top-level design, data standard management, data collection, data storage, and data application. These dimensions are derived through a combination of literature review and expert interviews. Subsequently, this study establishes a theoretical model titled “Data Governance Capability—Project Organization Resilience”, exploring the relationships among these data governance capability dimensions and their correlations with project organization resilience. Data were collected through 142 valid questionnaires from practitioners in the Chinese construction industry and analyzed by the Partial Least Squares Structural Equation Modeling (PLS-SEM) method. The results indicate a significant relationship between data governance capabilities and project organization resilience, contributing to the research on the antecedents of the latter. This study offers a scale for construction enterprises to systematically assess data governance capabilities and provides guidance on enhancing organization resilience by progressively nurturing these capabilities.

1. Introduction

In today’s VUCA (Volatility, Uncertainty, Complexity, Ambiguity) world, global supply chain restructuring, frequent public health crises, and disruptive digital technology innovations have intensified the dynamism and vulnerability of construction projects [1]. Existing studies indicate that adverse factors or crises, such as policy changes, project funding issues, market fluctuations, and major unforeseen events, can lead to difficulties in project initiation, supply chain disruptions, schedule delays, and cost overruns [2]. Resilience has gradually transcended the traditional scope of risk management, becoming a core strategic objective for project organizations to achieve counter-cyclical growth and sustainable development in times of crisis [3,4]. Therefore, enhancing the ability to respond to crises and disruptions is crucial in construction projects.

Existing studies have examined the antecedents of project organization resilience from various perspectives. First, the resource perspective highlights the role of strategic resource adequacy and allocation—such as financial assets, digital infrastructure, and human resources—in buffering against crises [5,6]. Second, psychological and cognitive perspectives stress leadership effectiveness and collective mindfulness in enhancing crisis response capacity, thereby mitigating the impact of adverse factors [7,8]. Additionally, the institutional perspective underscores the importance of social capital and relational networks in fostering rapid recovery after experiencing disruptions [9]. Finally, capability perspectives emphasize strong situational awareness and digital technology innovation as key enablers of resilience, by identifying and responding to risks at an early stage and leveraging new capabilities and adaptive strategies to absorb and capitalize on environmental changes [10,11].

While these studies provide valuable insights, two critical gaps persist. First, existing research predominantly focuses on resilience mechanisms at the enterprise or supply chain level [12,13,14]. Project organizations, as core units in the construction industry, lack a systematic examination of how to build their own resilience. This gap limits both theoretical and practical efforts to effectively enhance the adaptive capacity of project organizations. Second, although scholars widely acknowledge the enabling effect of digital technologies (such as BIM and the IoT) on resilience, the systematic impact of data governance capabilities remains understudied, especially since data governance has emerged as a hot topic now in the wave of digital transformation. For instance, while Hilkenmeier et al. (2021) found that data-driven decision-making frameworks enhance project organization agility [15], and Gölzer et al. (2017) linked optimized data processes to organizational stability [16], no study has holistically examined how structured data governance frameworks contribute to project organization resilience across the data lifecycle (planning–flow–application) from a capability-building perspective. Existing frameworks primarily focus on individual technological tools (e.g., data collection devices) or collaborative models (e.g., multi-party data-sharing platforms). To address those issues, this study aims to explore how construction enterprises enhance project organization resilience through a holistic data governance capability approach.

However, blindly implementing data governance may introduce challenges such as increased market volatility pressures, organizational rigidity, and insufficient strategic planning, which could negatively impact technological innovation and sustainable development. Existing data governance frameworks are often generic and lack specificity for construction enterprises. The sector-specific characteristics of data governance in construction (e.g., multi-entity collaboration and dynamic task environments) have not been fully incorporated into current frameworks, leading to misalignment between governance capabilities and resilience enhancement. Therefore, it is essential to identify the key dimensions of data governance tailored to the construction industry and develop a corresponding data governance capability scale.

The remainder of this study was organized as follows. Firstly, we defined and conceptualized data governance capabilities, and developed an initial measurement scale by conducting a systematic literature review. Secondly, we constructed a cross-level theoretical model linking data governance capabilities and project organization resilience, proposing specific hypotheses to elucidate the relationships between the dimensions of data governance capabilities and their impact on project organization resilience. Then, we refined the initial dimensions and indicators through expert interviews and validated the measurement framework using pre-test data and factor analysis, ultimately deriving the final data governance capability scale for construction enterprises. Finally, the data were analyzed by the PLS-SEM method to examine the proposed hypotheses. The Discussion and Conclusion sections show how this research enriches the existing literature and provide theoretical and practical guidance. The research design process diagram is shown in Figure 1.

2. Data Governance

2.1. Definition and Connotations of Data Governance

Data governance refers to an organization’s core capability to plan, control, and extract value from data assets through a systematic framework [17]. Its primary goal is to maximize the value of data, minimize risks and costs, and support the organization in decision-making and process management [18,19]. While there is no universally accepted definition in academia, theoretical advancements have progressively deepened its connotations. Early research took a narrow perspective, defining data governance as the control activities of data and its application processes, emphasizing technical measures to ensure data quality and security [20,21]. As digital transformation has advanced, the definition has evolved into a broader perspective, encompassing (1) authority and accountability over data assets [22], organizational structures and the distribution of decision rights in data processes [23], and (3) the alignment between data strategy and business goals [24]. The theoretical evolution of data governance can be categorized into three progressive stages, as summarized in

Table 1. Evolutionary stages of data governance.

Stage	Core Focus	Source
Technical Governance	Human–machine interaction and data quality	[20,25]
Organizational Design	Institutionalized structures and processes	[19,26,27]
Strategic Governance	Cross-functional collaboration and value creation	[24]

. This evolution reflects a paradigm shift from “managing data” to “governing through data”, where data reconfigure organizational decision-making.

Distinct from traditional data management, data governance emphasizes governance across the entire data lifecycle. It is an integrated governance framework that includes policies, processes, standards, and organizational structures [19]. Data governance extends beyond mere data collection and storage; it involves establishing decision-making authority and responsibilities, formulating standards and guidelines, and ensuring data quality, security, and compliance [19]. Existing research primarily focuses on four key perspectives, namely, the governance process, governance standards, constituent elements, and governance outcomes, which are summarized in

Table 2. Perspectives on data governance.

Perspective	Key Definitions	Source
Process	Cross-functional coordination of data lifecycle management that covers rule-making, division of responsibilities, organization and implementation, etc.	[28,29]
Standards	Their core objective is to improve data quality, which requires a governance framework integrating data infrastructure, standards, policies, etc.	[30]
Components	Data strategy, standardized processes, and governance mechanisms including structural, procedural, and relational mechanisms.	[21]
Outcomes	Operational optimization, cost reduction, strategic decision level, and competitiveness enhancement.	[26,31]

.

In the construction context, data governance needs to adapt to two key characteristics: project-driven nature and multi-stakeholder complexity. Since construction enterprises operate based on engineering projects, data originate from fragmented projects, necessitating governance capabilities that ensure quality, security, and the fluidity of data among various stakeholders across the full lifecycle (planning → flow → application). In summary, this study proposes that data governance in construction enterprises should encompass two main aspects: first, establishing digital platforms and standards for data transmission, storage, and utilization; second, optimizing workflows, redefining accountability, and aligning strategies to mitigate risks and unlock value in dynamic environments.

2.2. Identification of Data Governance Capability Dimensions

In construction enterprises, data governance capability is reflected not only in the effective management of the data lifecycle but also in the ability to extract value from data, enhance data-driven decision-making, and improve the efficiency of all stakeholders. Achieving effective data governance requires a high level of data governance capability maturity. To measure data governance activities, scholars have introduced the capability maturity model (CMM) into related research, leading to the development of the data governance capability maturity model. These models serve as essential tools for assessing an enterprise’s data governance capabilities by providing structured criteria for evaluating the development and management of software platforms. The following are the current representative models and their data governance capability dimensions (Figure 2):

Currently, there is no universally accepted classification of data governance capability dimensions in academia, so this study draws upon the most widely accepted data governance capability maturity models to establish a framework for data governance capability dimensions within the construction field. Given that data governance in construction management is still in its early stages, an overly detailed classification may hinder practical implementation. Therefore, based on value chain theory and data lifecycle theory, this study categorizes data governance into three stages: planning stage, flow stage, and application stage. By synthesizing the common dimensions across various data governance maturity models (Figure 2), we identify five key capability dimensions for data governance in the construction sector:

• Planning Stage: top-level design capability and data standard management capability.
• Flow Stage: data collection capability and data storage capability.
• Application Stage: data application capability.

Top-level design capability is the foundational guarantee for enterprises to implement data governance. It refers to an organization’s ability to formulate data governance strategies, establish governance teams, and develop governance frameworks based on digital transformation trends and internal resources [32]. Data standard management capability is critical for ensuring data consistency and accuracy. This capability involves developing standardized data definitions, classifications, and formatting guidelines. Data collection capability ensures that data are authentic, reliable, easily accessible, and timely. Effective data collection is crucial for high-quality data analysis and decision-making, as incomplete or inaccurate data can lead to inefficient management and operational risks. Data storage capability involves establishing standardized data storage protocols and equipping enterprises with specialized storage infrastructure to ensure data security, quality, and traceability. Enhancing this capability enables organizations to better protect and utilize their data assets, reducing data loss risks while improving data retrieval efficiency. Data application capability refers to an enterprise’s ability to analyze and leverage data for decision-making and business optimization.

After reviewing the content of the seven data governance capability maturity models mentioned earlier, it is evident that different countries and industries emphasize different aspects of data governance and adopt varying classification standards. There are significant differences among these maturity models in terms of capability dimensions and capability items, primarily reflected in the categories and number of capability dimensions and items they define. Additionally, data governance encompasses a wide range of aspects, with complex and extensive content, which further contributes to the variations among models. Moreover, differences in evaluation objectives and areas of focus across models further exacerbate these discrepancies. Despite their distinct approaches to capability domain classification, all these models commonly address key areas such as data strategy, data organization, data quality, data standards, data application, and data infrastructure.

Based on the findings above, data governance capabilities were consolidated and categorized according to the characteristics of the construction industry, resulting in five dimensions and 17 factors. The classification of data governance capability dimensions for construction enterprises is shown in

Table 3. Data governance capability dimension division.

No.	Stage	Capability Dimension	Factor	Source
1	Planning	Top-level Design	Data Strategy Planning	[24,33,34,35]
2			Data Governance System	[24,33,34,35]
3			Data Strategy Implementation	[24,34,35]
4			Data Governance Team	[24,33,34,35]
5		Data Standard Management	Data Standard Specifications	[19,35]
6			Data Standard System	[19,35]
7			Data Exchange Standards	[19,35]
8	Flow	Data Collection	Data Abundance	[19,24,35,36]
9			Data Accuracy	[19,24,35,36]
10			Data Timeliness	[19,24,35,36]
11		Data Storage	Data Storage Specifications	[33,36,37]
12			Data Storage Equipment	[33,36,37]
13			Data Storage Effectiveness	[33,36,37]
14	Application	Data Application	Data Operations and Maintenance	[24,33,34,36]
15			Data Analysis	[24,33,34,36]
16			Database Management	[24,33,34,36]
17			Data Exchange	[24,33,34,36]

.

3. Project Organization Resilience

Organizational resilience refers to an entity’s capacity to anticipate, prepare for, respond to, and adapt to both incremental changes and sudden disruptions while maintaining core functions and evolving in dynamic environments [38,39,40,41]. It encompasses the ability to absorb external shocks and achieve post-crisis equilibrium through adaptive learning [42,43]. However, despite the increasing attention given to the concept of resilience in the academic literature, the specific definition of project resilience remains largely unclear and ambiguous [39].

Current research in project management is primarily focused on traditional risk management [39]. While both risk management and project resilience share common antecedents such as disturbances, uncertainty, and change [44], risk management focuses on identifying and mitigating predictable threats. In contrast, project resilience acknowledges the presence of unknowns and emphasizes a proactive, adaptive approach to unforeseen challenges [45]. This perspective aligns with the dynamic capabilities theory, which posits that organizations must integrate, reconfigure, and utilize existing resources to maintain competitiveness in uncertain environments [46].

Project, as a temporary organizational entity, exhibit unique characteristics such as distinct lifecycle stages, high temporality, and inherent uncertainties. These features differentiate project organization resilience from traditional organizational resilience, necessitating a tailored approach that considers the transient and project-specific context [47]. Based on the existing representative definitions (

Table 4. Definitions of project organization resilience.

Reference	Key Definitions
Geambasu [49] (2011)	The ability of the project system to recover and continuously adapt to changes, as well as the ability to ensure the project operates effectively during disruptive events.
Giezen [50] (2015)	Conceptualization into prevention, reaction, and adaptation.
Turner and Kutsch [51] (2016)	The ability to detect and understand changes in the project environment, plan responses, minimize damage, and adapt to new realities.
Blay et al. [44] (2017)	The ability to respond to, prepare for, and mitigate the impacts of turbulent environments and project complexity, including proactivity, responsiveness, flexibility, and durability.
Zhang et al. [46] (2023)	The dynamic abilities of temporary project organizations to anticipate, respond to changes, adapt, and learn in changing environments to ensure projects can be effectively delivered, including anticipation, coping, and adaptation capabilities.

) and the capability-based perspective, project resilience can be conceptualized as a multifaceted capability that includes mitigating risks through anticipation, adapting to changes, and recovering from disruptions [48]. This framework of project organization resilience encompasses three core dimensions: defensive capability, responsive capability, and recovery capability.

Defensive capability refers to the proactive measures taken to anticipate and prepare for potential risks, including risk identification and mitigation strategies [49]. Rooted in anticipatory resilience, it aligns with the “sense–seize” phase of dynamic capabilities [52]. Responsive capability concentrates on problem-solving, resource reallocation, participant coordination, and the real-time handling of unexpected events or disruptions [46,53]. The aim is that project organizations can resume normal construction and operation in a timely manner after being disrupted [54]. These coping activities are typically constrained by available budget, resources (such as materials and human resources), and other factors. Recovery capability refers to post-crisis restoration through learning, adaptation, and process optimization [48], encompassing the dynamic ability of project organizations to self-adjust in anticipation of future disruptions.

4. Research Hypotheses

4.1. The Impact Between Dimensions of Data Governance Capabilities

Top-level design capability is fundamental to ensuring data quality and the realization of their value, and it serves as the strategic framework and essential safeguard for an organization’s data governance efforts [32]. Mature data governance top-level design capability can effectively address the issues of “what data to collect” and “what data to store”, while optimizing the allocation of an organization’s data resources. This enables better control over the quality and integrity of data, helping to avoid issues such as redundancy, errors, and inconsistencies. Some scholars argue that accurately understanding the framework and content of data governance is key to ensuring the accuracy, richness, and timeliness of data collection and storage, thus enabling the collected data to accurately reflect the actual situation of project development. This allows the organization to take a targeted approach during the data governance process, improving work efficiency and accelerating project progress.

H1a. 

The top-level design capability of data governance has a significant positive effect on data collection capability.

H1b. 

The top-level design capability of data governance has a significant positive effect on data storage capability.

Data standardization serves as the fundamental pillar of data quality and is crucial for ensuring data consistency, accuracy, and compliance. Well-defined data standards establish appropriate guidelines for data collection and storage, addressing the fundamental questions of “how data should be collected” and “how data should be stored”. This capability encompasses the establishment of data standards, including data definitions, classifications, and formatting guidelines. Such a standardized system facilitates the seamless integration, comparison, and analysis of data from diverse sources, formats, and quality levels while enabling the timely identification and regulatory handling of anomalous data. The absence of uniform data standards leads to various data quality issues, such as duplication, inaccuracy, and missing values, thereby compromising data reliability and accuracy and potentially resulting in misguided decision-making [55]. Furthermore, inconsistencies in data standards contribute to fragmented data storage and hinder horizontal data integration, forming “data silos” that obstruct data flow, introduce data risks and accountability challenges, and impede interdepartmental collaboration [55]. The harmonization of data structures, formats, and naming conventions reduces the workload associated with data transformation and cleansing, thereby enhancing data utilization efficiency.

H2a. 

Data standard management capability has a significant positive effect on data collection capability.

H2b. 

Data standard management capability has a significant positive effect on data storage capability.

High-quality data provide accurate information, making data application outcomes more reflective of reality and offering more valuable insights for enterprise decision-making. Enhancing data collection quality improves the efficiency of data analysis and shortens the analytical cycle. Data storage management capability involves establishing standardized data storage protocols, deploying specialized storage infrastructure, and ensuring data quality, security, and traceability. At the same time, it guarantees that various data models are consistently collected and converted into compatible formats for analysis [56]. During project implementation, effective data storage management enhances data utilization, reduces redundant data collection and processing, and consequently saves human, material, and financial resources. Furthermore, optimizing data storage structures and adopting advanced storage technologies can lower the costs associated with acquiring and maintaining data storage infrastructure. Ghasemaghaei and Calic argue that data storage serves as the foundation for ensuring data quality and plays a crucial role in enhancing the effectiveness of data application [57]. Therefore, strengthening data collection and storage management capability contributes to better protection and utilization of data assets.

H3a. 

Data collection capability has a significant positive effect on data application capability.

H3b. 

Data storage capability has a significant positive effect on data application capability.

4.2. The Impact of Data Governance Capability on Project Organization Resilience

Project organization resilience is manifested as the system’s adaptive capability to respond to uncertain shocks, a mechanism that can be explained by the dynamic capabilities theory and the resource-based view. As the final stage of data governance, data application capability systematically strengthens project organization resilience across three capability dimensions—defensive, responsive, and recovery—by supporting risk identification, resource allocation, and dynamic adjustments through the deep analysis of governed, high-quality project data [58].

Regarding defensive capability, data application capability enhances the defensive resilience of project organizations through risk anticipation and proactive preparedness [46]. Preparedness is a crucial prerequisite for coping with uncertainty. Before encountering adversity, data application capability is demonstrated by utilizing big data analytics to mine effectively managed historical project data, thereby identifying potential risks and providing preemptive decision-making support to mitigate the impact of adverse external factors [59]. Additionally, data-driven optimization improves information clarity, reduces environmental uncertainty, and sharpens market insights into market dynamics, thereby strengthening the ability to withstand external shocks [60,61].

Regarding responsive capability, data application capability enhances project organization resilience in three ways. First, when facing adversity, data-driven practices endow project organizations with greater flexibility, enabling them to rapidly identify innovation opportunities and adapt to environmental changes [62]. Second, the resource structure of construction projects is often not effectively utilized in risk response [46]; especially in severe crises, an organization lacking sufficient response capability may experience project interruption, cooperation failure, or even negative outcomes [63]. Under resource constraints, data analytics enable precise resource allocation, improving crisis response efficiency [46]. Third, data-sharing mechanisms break down supply chain data barriers by synchronizing cross-team decision-making through real-time dashboards, reducing communication costs, and expediting crisis resolution [64,65].

Regarding recovery capability, data application capability fosters recovery resilience through knowledge integration and innovation. Post-crisis, project organizations leverage data governance to integrate internal and external resources and, through exploratory and exploitative innovation, absorb new knowledge and seek development opportunities [59]. Data-driven decision systems optimize survival capability under resource constraints, for example, by recording crisis response experiences in knowledge repositories to provide a foundation for subsequent project recovery [50]. Such adaptive learning ensures continuous improvement in volatile environments [38].

H4a. 

Data application capability has a significant positive effect on project organization defensive capability.

H4b. 

Data application capability has a significant positive effect on project organization responsive capability.

H4c. 

Data application capability has a significant positive effect on project organization recovery capability.

The model of data governance capabilities and project organization resilience is depicted in Figure 3.

5. Methods

The methods of this study follow a four-stage workflow, which is depicted in Figure 4. Stage 1 establishes the items of the scale and lays a solid foundation for follow-up research. Stage 2 ensures rigorous data collection through iterative scale refinement and pilot testing. Stage 3 validates the measurement model’s reliability and construct validity, while Stage 4 employs PLS-SEM to test hypotheses and control for confounding factors.

5.1. Sample

This study collected empirical data from professionals in the Chinese construction industry through an online survey distributed via the Questionnaire Star platform between December 2023 and March 2024. After a rigorous screening and sorting process, a total of 145 valid questionnaires were collected. To ensure the accuracy and reliability of the research data, two criteria were used to determine invalid responses: (1) selecting the same option for multiple consecutive questions and (2) choosing mutually exclusive options within the same scale. Regarding the selection of respondents’ job categories, this study specifically targeted employees of general contracting companies and those involved in general contracting projects to ensure the focus remained on these enterprises or projects. Additionally, the target respondents were required to have a deep understanding of data governance, BIM, and digitalization technologies. The majority of the respondents had over five years of experience in the construction industry, with a strong understanding of construction projects and the ability to independently manage medium-to-large-scale projects.

Based on the above criteria, a final total of 142 valid questionnaires were obtained, resulting in a high validity rate of 97.9%. These data provide a solid foundation for subsequent statistical analysis, ensuring the rigor and reliability of the research results.

Table 5. Description of sample characteristic distribution.

Variable	Option	Frequency	Percentage
Enterprise Type	State-owned enterprise	84	59.20%
	Joint-stock enterprise	4	2.80%
	Private enterprise	39	27.50%
	Other	15	10.60%
Project Type (Multiple Choice)	Housing construction	97	68.20%
	Road and bridge	59	41.50%
	Port/transportation	14	9.90%
	Energy	22	15.50%
	Municipal	55	38.70%
	Telecommunications	10	7.00%
	Industry	10	7.00%
	Other	27	19.00%
Gender	Male	90	63.40%
	Female	52	36.60%
Age	<25	37	26.10%
	25–30	54	38.00%
	31–35	21	14.80%
	36–40	21	14.80%
	>40	9	6.30%
Years of Work Experience	<1	29	20.40%
	1–5	60	42.30%
	6–10	26	18.30%
	>10	27	19.00%
Education	Below undergraduate	2	1.40%
	Undergraduate	70	49.30%
	Master’s degree	67	47.20%
	Doctorate	3	2.10%
Position	Senior manager	5	3.50%
	Mid-level manager	34	23.90%
	Junior-level manager	51	35.90%
	Staff member	46	32.40%
	Other	6	4.20%

presents detailed background information on the projects and respondents.

5.2. Measurement

To obtain a more accurate, scientific, and reliable data governance capability dimension list, the initially identified dimensions of data governance needed further screening and refinement to align with the actual conditions of construction enterprises and projects, as well as meeting the requirements for feasibility studies. Therefore, expert interviews were conducted using the “Expert Survey on Data Governance Capability Dimensions for Construction Enterprises” to solicit professional opinions. Expert interviews are a widely used analytical method in social science research, known for their simplicity and effectiveness in overcoming the limitations of group decision-making, ensuring that important factors are not overlooked during the decision-making process. Based on the preliminary identification of the data governance capability dimensions through literature search, this study used expert interviews to further screen and refine the dimensions, and then revised and categorized all data governance capabilities according to the experts’ suggestions. This process resulted in the final list of data governance capability dimensions, which helps to address the gaps and omissions in the literature search, laying a solid foundation for the establishment of the structural equation model in the next chapter.

The study selected 12 experienced industry experts as participants in the survey. These experts have been engaged in the relevant fields for over five years, possessing strong professional backgrounds and extensive practical experience. Among them are 3 university professors with high academic prestige and influence, 5 general contracting project managers with rich hands-on experience and outstanding leadership skills, and 4 professionals from the digital transformation center of general contracting enterprises, who possess forward-thinking approaches and expertise in digital transformation and innovation. The participation of these experts will provide valuable industry insights and professional recommendations for this study.

Based on expert interview feedback, modifications were made to the initial data governance capability dimensions. Since “data standard specification” and “data exchange standard” overlapped, “data standard specification” was revised to “project-level data standard specification”, and “data exchange standard” was modified to “cross-enterprise and departmental data exchange standard”. Additionally, considering the widespread application of visualization data platforms in construction project management, data visualization capability was deemed essential. Therefore, “data visualization” was added under the data application capability dimension.

The final list of data governance capability dimensions for construction enterprises (

Table 6. Final version of the construction enterprise data governance capability dimensions.

Dimension	Factor	Description
Top-Level Design Capability	Data Strategy Planning	Defining the vision and objectives of data management activities; during the project initiation phase, strategic planning documents for data governance are developed.
	Data Governance System	Establishing a corresponding data governance system, with periodic reviews and updates, to ensure standardized execution of data governance tasks.
	Data Governance Organization	Constructing an organizational framework that complements the data governance system; this framework clarifies responsibilities and ensures efficient internal communication.
Data Standard Management Capability	Project-Level Data Standard Specifications	Establishing project-level data standards to ensure consistency across different project phases, such as BIM data standards for design and construction stages.
	Data Standard System	Developing data standard systems, including data definitions, classifications, and formats.
	Cross-Enterprise and Departmental Data Exchange Standards	Adopting industry or national standards to regulate internal and external data exchange.
Data Collection Capability	Data Richness	Collecting data from multiple sources to ensure data comprehensiveness.
	Data Accuracy	Establishing a data quality review system to ensure the accuracy of data reflecting the actual project conditions.
	Data Timeliness	Periodically updating data to maintain data quality and timeliness.
Data Storage Capability	Data Storage Standards	Establishing data storage standards that define data storage processes, methods, and naming conventions.
	Data Storage Equipment	Equipping dedicated data storage devices and ensuring timely project data archiving and backup to facilitate data traceability.
	Effectiveness of Data Storage	Regularly inspecting data storage effectiveness to ensure data quality and security.
Data Application Capability	Data Analysis	Providing data decision support for project implementation by conducting exploratory data analysis and leveraging digital platforms for project management and resilience assessment.
	Database Management	Establishing project databases to facilitate data retrieval and support decision-making.
	Data Exchange	Enabling data sharing through data platforms to enhance project collaboration.
	Data Visualization	Utilizing data platforms to monitor project construction in real time.

) classifies data governance capabilities into five primary dimensions and sixteen secondary dimensions. The specific categorization is as follows:

• Top-Level Design Capability: Includes three factors—data strategy planning, data governance system, and data governance organization.
• Data Standard Management Capability: Includes three factors—project-level data standard specification, data standard system, and cross-enterprise and departmental data exchange standard.
• Data Collection Capability: Includes three factors—data richness, data accuracy, and data timeliness.
• Data Storage Capability: Includes three factors—data storage specification, data storage equipment, and data storage effectiveness.
• Data Application Capability: Includes four factors—data analysis, database management, data sharing, and data visualization.

After refining the dimensions of data governance capability, a scale for measuring data governance capability in construction enterprises was developed. Based on the theoretical model, the scale items were formulated by extensively referencing the existing literature and the aforementioned data governance capability maturity model classifications [28,33,36,66]. Considering the actual operational conditions and characteristics of enterprises in the construction sector, the relevant scale was carefully adjusted and optimized, leading to the development of a preliminary survey questionnaire for measuring data governance capability. To validate the effectiveness of the questionnaire, a pilot survey was conducted among 18 project managers, confirming its applicability for large-scale research. Respondents were asked to answer the questions based on their most recently completed construction project. The first part of the questionnaire included control variables and background information on the respondents and their projects. Excluding this section, all items were measured using a five-point Likert scale (e.g., 1 = Strongly Disagree; 5 = Strongly Agree).

The measurement items for data governance capability are presented in

Table 7. Data governance capability measurement items.

Dimension	Code	Measurement Items
Top-Level Design Capability	DC1	The company has formulated a comprehensive data governance strategy (including vision, goals, and principles).
	DC2	The company has established a sound data governance institutional system to ensure sustainable data management.
	DC3	The company has formed a data governance team, clarifying relevant roles, responsibilities, and workflows.
Data Standard Management Capability	BZ1	The company has established project-level data standards.
	BZ2	The company has formulated a data standard system that includes data definitions, classifications, and formats.
	BZ3	The company adopts industry or national standards to regulate data exchange within and outside the team.
Data Collection Capability	CJ1	The company can collect data from multiple channels to ensure data richness.
	CJ2	The company has established a data quality review system to ensure data accuracy in reflecting project conditions.
	CJ3	The company can ensure timely data updates.
Data Storage Capability	CC1	The company has established data storage specifications.
	CC2	The company is equipped with dedicated data storage devices and ensures timely project data archiving and backup.
	CC3	The company regularly inspects the effectiveness of data storage to ensure data quality and security.
Data Application Capability	YY1	The company can allocate data according to its strategy and business needs.
	YY2	The company has established a database to facilitate data retrieval and support decision-making.
	YY3	Departments within the company can share data through data platforms, enhancing collaboration.
	YY4	The company can observe project construction status in real time through data platforms.

.

This study identifies three dimensions of project organization resilience through a systematic literature review, namely, defensive capability, responsive capability, and recovery capability, and provides an in-depth analysis of their connotations. Defensive capability refers to an organization’s ability to anticipate unexpected events and proactively prepare for risks. The measurement items in this section are primarily based on the theoretical research of Duchek et al. [38]. Responsive capability refers to the ability to acknowledge problems and promptly formulate and implement solutions in response to internal and external changes. The measurement items in this section mainly draw from the research findings of Home and Orr [67], Kendra and Wachtendorf [68], and Lee et al. [7]. Recovery capability refers to an organization’s ability to recover from crises in a short period while learning from experiences to lay a solid foundation for future projects. The measurement items in this section are primarily based on the research of Sanchis et al. [69].

The measurement items for project organization resilience are presented in

Table 8. Project organization resilience measurement items.

Dimension	Code	Measurement Items
Defensive Capability	FY1	The project organization has a clear understanding of potential project risks.
	FY2	The project organization can continuously monitor the project execution process to issue warnings for emerging problems.
	FY3	The project organization has developed an emergency response plan and conducts regular drills for emergency situations.
Responsive Capability	XY1	When a risk occurs, the project organization can coordinate internally to ensure entering a systematic response state.
	XY2	When a risk occurs, the project organization can quickly develop response plans and take action.
	XY3	The project organization can quickly identify and assess different types of crises.
Recovery Capability	HF1	The project organization has established a formal emergency team or group to mitigate crises.
	HF2	The project organization can recover from sudden incidents or risks with minimal time and cost.
	HF3	The project organization can learn from past crises, absorb experience, and lay a foundation for future project construction.

.

According to previous research, three control variables that are likely to influence project organization resilience are considered to minimize issues related to omitted variable bias [70]. First, human resource, the leadership and competence of core employees, contributes to organizational resilience in crisis response [5]. Human resource was measured by the following single item: “The enterprise possesses leaders with digital strategic vision and employees proficient in applying digital technologies in their work”. Second, technical resource may empower project organizational resilience in enabling virtual communication and remote access to resources [71]. Technical resource was measured by the following single item: “The enterprise has comprehensive software resources, hardware facilities, and physical spaces related to the application of digital technologies”. Finally, Organization resource, which in some sense reflects the company culture, was measured by the following single item: “The enterprise has established corresponding organizational structures and institutional systems for implementing digital technologies”.

6. Data Analysis and Results

6.1. Reliability and Validity Analysis

Reliability and validity testing ensures the rationality of questionnaire design and the accuracy of sample data in reflecting variable responses.

First, principal component analysis (PCA) was employed to conduct exploratory factor analysis (EFA) to assess the factor structure and item design of the questionnaire. The Kaiser–Meyer–Olkin (KMO) test and Bartlett’s sphericity test were used to determine whether each dimension was suitable for factor analysis. The threshold criteria were set as KMO > 0.5 and Bartlett’s test p < 0.05 to confirm the appropriateness of factor analysis. Cronbach’s α coefficient was also examined, with α > 0.7 considered indicative of good reliability. The reliability test was performed using SPSS 26.0, yielding an overall KMO value of 0.944, indicating a good model fit for factor analysis. The Bartlett test statistic was significant at the p < 0.001 level, further validating the suitability of factor analysis. As shown in

Table 9. Reliability and validity test results.

Construct and Measure Items	SFL	Cronbach’s α	CR	AVE
Top-Level Design Capability		0.878	0.927	0.810
DC1	0.903
DC2	0.912
DC3	0.886
Data Standard Management Capability		0.836	0.902	0.753
BZ1	0.899
BZ2	0.840
BZ3	0.864
Data Collection Capability		0.866	0.918	0.788
CJ1	0.878
CJ2	0.900
CJ3	0.884
Data Storage Capability		0.878	0.925	0.805
CC1	0.892
CC2	0.861
CC3	0.937
Data Application Capability		0.932	0.951	0.830
YY1	0.928
YY2	0.886
YY3	0.913
YY4	0.917
Defensive Capability		0.793	0.879	0.708
FY1	0.810
FY2	0.843
FY3	0.869
Responsive Capability		0.889	0.931	0.819
XY1	0.926
XY2	0.889
XY3	0.898
Recovery Capability		0.863	0.916	0.784
HF1	0.853
HF2	0.888
HF3	0.915

, the Cronbach’s α coefficients for DC, CJ, BZ, CC, YY, FY, XY, and HF were 0.878, 0.866, 0.836, 0.878, 0.932, 0.793, 0.889, and 0.863, respectively (all exceeding 0.7), demonstrating the high internal consistency and strong reliability of the questionnaire. After examining the factor loadings for each item, no cross-loading issues were identified, so no items needed to be eliminated.

Second, confirmatory factor analysis (CFA) was conducted using Amos 24.0 to validate the factor structure and item design established in the EFA. Figure 5 shows the first-lever factor analysis results of the scale. The standardized factor loadings (SFLs) for all items exceeded 0.7, the average variance extracted (AVE) for all variables was greater than 0.7, and the construct reliability (CR) values ranged between 0.879 and 0.951, indicating strong convergent validity and composite reliability. Additionally, discriminant validity analysis confirmed significant distinctions among different variables, demonstrating that the indicators effectively differentiate between constructs. As shown in Table 9, all variables exhibited significant correlations, and the square root of each variable’s AVE was greater than its off-diagonal correlation coefficients, supporting good discriminant validity. Furthermore, all heterotrait–monotrait (HTMT) values were below the 0.90 threshold [72].

Finally, the fitting indices for the first-level factor analysis are shown in

Table 10. Fitting indices of confirmatory factor analysis.

Index	χ2/df	RMR	CFI	IFI	TLI
Standard	<3	<0.05	>0.9	>0.9	>0.9
Value	1.950	0.031	0.928	0.930	0.913

. According to Table 10, the fit values obtained from the first-level confirmatory factor analysis (χ2/df = 1.950 < 3; RMR = 0.031 < 0.05; CFI = 0.928 > 0.9; IFI = 0.930 > 0.9; TLI = 0.913 > 0.9) indicate an acceptable level of agreement, and the CFA results indicate that the construct validity of the scale is appropriate.

6.2. Hypothesis Testing

This study employed Smart PLS 4.0 to test the research hypotheses using the Bootstrapping algorithm with 5000 resamples. As shown in

Table 11. Results for hypothesis testing.

Hypothesis	Path	β	t	CI		p	Correlation
				2.5%	97.5%
H1a	DC → CJ	0.445	5.108	0.281	0.624	0.000	+
H1b	DC → CC	0.242	3.521	0.110	0.380	0.000	+
H2a	BZ → CJ	0.456	4.952	0.267	0.624	0.000	+
H2b	BZ → CC	0.655	10.096	0.524	0.780	0.000	+
H3a	CJ → YY	0.379	4.070	0.202	0.567	0.000	+
H3b	CC → YY	0.537	5.646	0.343	0.712	0.000	+
H4a	YY → FY	0.677	11.765	0.554	0.779	0.000	+
H4b	YY → XY	0.677	11.433	0.550	0.780	0.000	+
H4c	YY → HF	0.657	9.872	0.513	0.773	0.000	+

Note: DC = top-level design capability; BZ = data standard management capability; CJ = data collection capability; CC = data storage capability; YY = data application capability; FY = defensive capability; XY = responsive capability; HF = recovery capability.

, the path coefficients (β) from top-level design capability to data collection capability and data storage capability were 0.445 and 0.242, respectively, both significant at the 0.001 level. This indicates that top-level design capability positively influences data collection and storage capabilities, supporting H1a and H1b.

Similarly, the path coefficients (β) from data standard management capability to data collection capability and data storage capability were 0.456 and 0.655, respectively, both significant at the 0.001 level. This confirms the positive impact of data standard management capability on data collection and storage capabilities, supporting H2a and H2b.

The path coefficients (β) from data collection capability and data storage capability to data application capability were 0.378 and 0.538, respectively, both significant at the 0.001 level. This demonstrates that data collection and storage capabilities positively influence data application capability, supporting H3a and H3b.

Finally, the path coefficients (β) from data application capability to project organization resilience (defensive capability, responsive capability, and recovery capability) were 0.677, 0.677, and 0.657, also significant at the 0.001 level. This confirms that data application capability has a positive impact on project organization resilience, supporting H4.

With respect to the control variables (human resource, technical resource, and organization resource), our results illustrate that only human resource has a positive effect on project organization resilience (β = 0.199; p = 0.019) and other variables do not have a significant effect. Furthermore, when control variables are included in the analysis, they do not significantly alter the main results. This suggests that the primary relationships observed in our study remain robust, regardless of the inclusion of these control variables.

The results of the hypothesis testing are shown in Figure 6.

7. Discussion

This study aims to explore the influence of data governance capability on the resilience of construction project organizations. To achieve this, we first identified the key dimensions of data governance capability in construction enterprises and developed corresponding measurement scales. The findings not only validate the rationality of these dimensions but also reveal their inter-connections, offering a new theoretical perspective on the application of data governance capability in the field of construction management.

7.1. Dimensions and Hierarchical Relationships of Data Governance Capability

This study introduces five core dimensions of data governance capability in the field of construction management: top-level design, data standard management, data collection, data storage, and data application. More importantly, based on the implementation process of data governance, these dimensions are categorized into three stages, namely, planning, flow, and application, and their inter-connections are verified. The results indicate that these dimensions influence each other, addressing a gap in existing research. Previous studies have primarily enumerated different dimensions of data governance [56,73,74], but have lacked an in-depth discussion of how these dimensions interact with one another.

Consistent with the IBM data governance maturity model and the findings of Mahanti (2021) [66], the validation of H1 and H2 demonstrates that top-level design and data standard management (planning stage) positively influence data collection and data storage capabilities (flow stage). Khatri and Brown (2010) [19] also proposed a similar viewpoint, stating that an effective top-level design in data governance can optimize data collection and storage management processes, thereby improving data quality and usability. However, unlike traditional IT governance frameworks, this study emphasizes the strategic role of data governance, arguing that top-level design not only affects the efficiency of data management but also has profound implications for an enterprise’s long-term development and risk resilience, which is consistent with the arguments proposed by Alhassan et al. (2018) [18]. Different from previous studies, this study identifies data governance dimensions more suited to construction projects through interviews with industry experts, thereby making a significant contribution to the existing literature. Furthermore, this study refines the influence pathways of data standard management, emphasizing its role in enhancing data fluidity and integrated management. As the core of the data governance system, data standard management ensures data standardization, accuracy, and consistency, which are crucial for effective data governance.

H3 is supported by the results, indicating that the circulation stage of data governance (data collection and storage) positively influences the application stage. This finding aligns with the study of Alhassan et al. (2018) [18], which suggests that high-quality data collection and storage management are prerequisites for maximizing data application value. Furthermore, this study confirms that, in the context of construction projects, data integrity and traceability enhance the effectiveness of data analysis, providing organizations with more accurate decision-making support.

7.2. The Impact of Data Governance Capability on Project Organization Resilience

This study innovatively establishes the path mechanism through which data governance capability influences project organization resilience. Previous research has primarily explored the potential impact of digital technology [11,71,75], multi-party collaboration [76], data analytics [77], or data management [78] from a singular perspective. Among them, the role of BIM, IoT, and other digital technologies in project organizational resilience is most often mentioned. These technologies provide the necessary tools and infrastructure for data governance, but their effectiveness hinges on alignment with governance strategies. Technology alone cannot address the systemic challenges of data governance—such as stakeholder misalignment or fragmented accountability—which require institutionalized processes and strategic oversight. This study’s framework thus prioritizes governance capabilities over tool-specific functionalities, ensuring adaptability across heterogeneous technological environments. By consolidating these fragmented perspectives into a unified data governance capability framework, this study fills this gap and reveals how data governance capability, as a holistic concept, enhances project organization resilience by strengthening its defense, response, and recovery capabilities.

The validation of H4 confirms that data governance capability has a positive impact on project organization resilience. Burnard et al. (2018) emphasized that the core of organization resilience lies in information integration and dynamic adaptability [79]. This study further demonstrates that improving data application capability enhances a construction project organization’s ability to manage risks and foster innovation. Particularly in project management, data application capability not only optimizes internal management but also promotes cross-project knowledge sharing, thereby improving a project’s ability to recover from crises. This finding offers new managerial insights for construction enterprises seeking to enhance organization resilience in uncertain environments.

8. Conclusions

This study aims to explore the mechanisms through which data governance capability influences organization resilience in construction projects. The findings validate that data governance enhances project organization resilience—including defensive capability, responsive capability, and recovery capability—via strengthened data application capability. Furthermore, the five dimensions of data governance capability do not exist in isolation but follow an inherent logic of data flow, forming an integrated system: top-level design provides a guiding framework for data collection and storage, data standard management regulates data collection processes and storage formats, high-quality data collection ensures the value of stored data, and effective data storage ultimately supports data application. These insights provide a novel pathway for improving construction enterprises’ capacity to navigate uncertainties and crises.

8.1. Theoretical Contributions

First, systematizing data governance capability dimensions: This study consolidates fragmented data governance dimensions into an integrated framework with measurable indicators (five dimensions across three stages), thereby addressing conceptual ambiguity in the construction data governance literature and providing theoretical foundations for systematic governance system development.

Second, clarifying hierarchical interdependencies of data governance capability: By investigating dynamic influence mechanisms among capability dimensions through a value-chain logic (planning → flow → application), this study advances beyond static technical interpretations of data governance. These empirically verified sequential relationships offer actionable phased pathways for enhancing capabilities from strategic planning to operational implementation.

Third, establishing the “Data Governance Capability–Project Organization Resilience” theoretical framework: Through theoretical integration, this study positions data governance capability as a critical antecedent of project organization resilience. This framework equips project managers with concrete operational levers while enriching interdisciplinary research at the nexus of data governance and project resilience.

8.2. Managerial Implications

First, establishing a systematic data governance framework: Enterprises should strengthen the top-level design of data governance by defining clear data governance strategies and aligning data management with overall corporate objectives, which involves establishing cross-departmental data governance teams and forming dedicated committees with explicit roles and responsibilities. Additionally, regular reviews should be implemented and efforts made to promote data standardization to develop industry-wide data specifications, enhancing data fluidity and sharing efficiency.

Second, enhancing data quality management to improve data usability: Enterprises should refine data standard management policies, optimize data collection and storage processes, and ensure data integrity, accuracy, and security. For instance, automated data collection tools can be leveraged to reduce human errors, while blockchain and distributed storage technologies can be adopted to enhance data traceability.

Third, developing a resilience management system driven by data application: The true value of data lies in their application. Construction enterprises should increase investment in data analytics and visualization technologies to drive data-informed decision-making, for example, during the defensive stage, using historical data patterns for risk prediction systems; during the responsive stage, establishing real-time monitoring dashboards for crisis management; and during the recovery stage, documenting crisis response lessons in knowledge repositories.

8.3. Limitations and Future Research Directions

First, the data collection in this study is limited to professionals in China’s construction industry, without considering governance differences in international projects or multicultural contexts. Data governance policies vary significantly across regions, influencing the capabilities of construction enterprises differently. Future research could enhance sample size and diversity, expanding into cross-national comparative studies. Second, while this study provides a framework for understanding the interplay between data governance and project organization resilience, it does not explicitly incorporate external environmental factors such as market conditions, regulatory policies, and industry competition. Future studies could introduce variables such as environmental uncertainty and project complexity as moderating factors to further examine how external conditions impact the relationship between data governance capability and project organization resilience.