Identifying the Factors Hindering Stakeholder Management in Construction with Structural Equation Modeling

Abstract

Effective stakeholder management is essential in construction projects, but numerous context-specific challenges often hinder its implementation. This study investigates the factors that limit stakeholder management in Türkiye’s construction industry through a structured, multi-stage analytical approach. A systematic literature review first identified 69 stakeholder management challenges (SMCs). A questionnaire administered to 164 construction professionals was then analyzed using the normalized mean value (NMV) approach, which identified 53 critical challenges (CCs). To reduce the dimensionality of the 53 CCs, exploratory factor analysis (EFA) was conducted, resulting in four overarching factors: (1) weak planning, coordination, and implementation deficiencies; (2) institutional and operational weaknesses; (3) communication problems; and (4) legal regulations, bureaucratic barriers, and ethical issues. Finally, structural equation modeling (SEM) was applied to highlight the effect sizes of these factors in stakeholder management, rather than to perform predictive modeling. The results show that institutional and operational weaknesses and communication problems exert the strongest negative influences. By clearly linking the 53 CCs with four higher-level factors, this study provides a coherent analytical structure and a robust methodological basis for understanding the barriers to effective stakeholder management. The findings offer actionable insights for construction practitioners and policymakers seeking to enhance coordination, communication, and governance mechanisms in complex project environments.

1. Introduction

A stakeholder is any individual, group, or institution that can influence, or is influenced by, a project’s goals, processes, or outcomes [1]. Stakeholder management is the systematic process of engaging stakeholders to identify, negotiate, and achieve social, environmental, and economic objectives through active participation [2]. As Bryson [3] notes, this involves strategically managing relationships by analyzing needs and expectations, establishing communication channels, and aligning interactions with the organization’s mission.

The foundations of stakeholder management are commonly linked to Freeman’s [4] seminal work, “Strategic Management: A Stakeholder Approach”. At its core, stakeholder management focuses on the complex relationships between organizations and the various groups that can influence—or be influenced by—a project. These interactions may create both opportunities and challenges for the project and the parties involved. Consequently, the purpose of stakeholder management is to help organizations recognize, assess, interpret, and appropriately engage with their stakeholders [5]. Although originally rooted in business management, stakeholder theory has been increasingly adopted in other disciplines, including construction management. However, because construction projects have distinctive characteristics and operational processes, there is a need to establish principles tailored specifically to managing stakeholders in the construction context.

Stakeholder theory suggests that organizations ought to consider the interests and expectations of all stakeholder groups, rather than focusing solely on shareholders. Adopting this perspective encourages a more comprehensive understanding of project outcomes and helps project teams manage the multifaceted dynamics that arise in stakeholder relationships [6]. When executed effectively, stakeholder management minimizes risks, builds a competitive advantage, and supports long-term growth [7]. This is particularly relevant in the construction industry, where projects are complex, involve multiple parties, and are subject to uncertainty [8]. Construction significantly contributes to the global economy through a diverse range of projects [9]. However, with large budgets, extended timelines, and multiple stakeholders, projects remain highly dynamic [8]. Despite stakeholders recognizing the importance, ineffective management remains a common issue [10]. Neglecting stakeholder needs can lead to cost overruns, delays, disputes, and social tensions [11].

Research indicates that poor communication, weak information exchange, and inadequate conflict or stakeholder management significantly contribute to project failure [12]. Diverging interests often intensify conflicts [11,13]. According to PMBOK, managing expectations is essential for ongoing progress. Identifying barriers is, therefore, key to success. Effective engagement enhances efficiency and promotes sustainable, resilient practices across the industry [14]

Scholars have extensively studied stakeholder management; however, research indicates that conditions in developing countries are particularly challenging. Notable differences are evident between developed and developing regions, driven by variations in research focus, methodologies, challenges, and practices.

In developed countries, research indicates the use of structured, formal approaches supported by empirical evidence. Yang et al. [15] emphasize the importance of systematic frameworks for stakeholder engagement, while Taimu et al. [16] note that robust practices are key to driving project success.

In contrast, developing countries apply weaker principles. Ola-awo et al. [17] highlight recurring time and cost overruns, disputes, and distrust. Eyiah-Botwe et al. [18] found that critical success factors in Ghana remain unaddressed. Kululanga [19] notes limited research on capacity building in Sub-Saharan Africa, while Onososen et al. [20] show barriers to adopting digital technologies that complicate interactions. Socio-cultural issues and fragmented projects add further complexity, which is notably absent in developed contexts.

Identifying critical factors is especially important for developing countries, where construction plays a significant role in driving economic growth [21]. The relationship between the construction sector’s share in Gross Domestic Product (GDP) and a country’s level of economic development has been widely discussed by Bon [22]. Bon [22] indicated that in low-income economies, the construction sector initially represents a relatively small share of GDP but expands rapidly during periods of industrialization and infrastructure development. As countries transition to higher-income stages, this share gradually stabilizes and eventually declines, reflecting a shift toward service-oriented and technology-driven economic structures. This non-linear pattern highlights the evolving role of construction activity across different stages of economic development.

In Turkey, construction plays a vital role, particularly in large-scale infrastructure such as airports, highways, bridges, dams, and power systems [23]. It can contribute up to 30% of GDP [24], making it a cornerstone of growth. However, persistent challenges, such as weak coordination, fragmented communication, bureaucratic inefficiencies, and gaps in professional competence, continue to obstruct effective stakeholder collaboration. Therefore, identifying barriers to sustainable construction is crucial. Although numerous studies [25,26,27] have discussed individual stakeholder-related challenges, no prior research has systematically integrated evidence from the literature with practitioner-based prioritization conducting normalized mean value analysis (NMV) and multistep factor–analytic validation performing exploratory factor analysis (EFA), confirmatory factor analysis (CFA) and structural equation modeling (SEM) to develop an empirically supported model of critical stakeholder management factors in the Turkish construction context.

Stakeholder management is essential to the success of construction projects, where multiple actors with diverse expectations must collaborate amid high uncertainty and complexity. When not managed effectively, stakeholder-related issues contribute to delays, cost overruns, disputes, coordination failures, and reduced project performance. Although the broader topic of stakeholder management has been widely examined, there is limited understanding of the barriers to stakeholder engagement—particularly in developing countries, where institutional, operational, and regulatory constraints are more pronounced.

This gap raises several vital research questions (RQs). First, although numerous issues affecting stakeholder management have been discussed in the literature, it is unclear which potential barriers are most relevant in the Turkish context (RQ1). Second, even when multiple challenges are identified, not all exert the same level of influence, making it necessary to determine which barriers are truly critical (RQ2). Third, the literature rarely consolidates these challenges into a meaningful set of underlying dimensions, leaving open the question of which critical factors structurally hinder effective stakeholder management (RQ3). Finally, while analytical tools such as structural equation modeling (SEM) are used descriptively, little is known about the relative effect sizes of these critical factors on stakeholder management performance (RQ4).

Accordingly, this study seeks to address these gaps by integrating evidence from a systematic literature review, a nationwide survey, NMV, EFA, CFA, and SEM. The following research questions guide the investigation:

RQ1: What are the potential barriers to effective stakeholder management?

RQ2: What are the critical barriers to stakeholder management?

RQ3: What are the critical factors hindering effective stakeholder management?

RQ4: What are the effect sizes of the critical factors influencing stakeholder management?

In this study, “critical factors” refer to the stakeholder management challenges that have the strongest and most consistent influence on project outcomes. These factors are identified through a two-stage process. First, all challenges are evaluated using the NMV analysis, and only those scoring above NMV > 0.5 are retained. This cutoff is widely used in construction-management and perception-based studies [28,29] because NMVs greater than 0.5 indicate that respondents rated an item above the neutral/average level of influence. Several prior studies have applied the same criterion when filtering perception-based variables before factorial analyses or modeling—e.g., Zhao et al. [28] and Liao and Teo [29]—all of whom classified items with NMV > 0.5 as high-priority issues. Following these established precedents ensures methodological consistency and allows the exclusion of low-impact items that introduce statistical noise into subsequent EFA, CFA, and SEM.

Second, the retained items are grouped and validated through exploratory and confirmatory factor analysis, which reveal the underlying dimensions that consistently shape stakeholder management performance. The factors emerging from this combined statistical- and perception-based process are therefore classified as “critical” because they represent the most significant and empirically supported barriers within the dataset.

By answering these questions, this study provides a comprehensive, empirically grounded understanding of the obstacles to stakeholder management in Türkiye’s construction industry. The findings of this study may be generalized to several developing countries and offer practical guidance for improving coordination, communication, and organizational effectiveness, while also contributing conceptual clarity to the broader stakeholder management literature.

2. Existing Studies on Stakeholder Management in the Construction Industry and Literature Gaps

A significant amount of research has explored stakeholder management in the construction industry. Yang et al. [15] conducted a systematic review and provided suggestions for future research, while Xia et al. [30] combined risk and stakeholder management using a similar approach. Yang et al. [31] identified gaps in earlier studies using literature reviews and interviews. Mok et al. [32] focused on mega-projects, and Oppong et al. [1] examined performance attributes. Prebanić & Vukomanović [33] connected stakeholder management with digital transformation. Yang and Shen [34] created a framework based on interviews.

Beyond reviews, Frempong-Jnr et al. [35] studied the impact of stakeholder management on construction waste management through a questionnaire and quantitative analysis. Yang et al. [36] and Mashali et al. [37] examined critical success factors in Hong Kong and Qatar, respectively, using surveys and descriptive analyses.

The stakeholder management challenges within the construction industry in developing countries have been analyzed in recent literature, revealing common themes that underscore the complexity and uniqueness of these environments. A predominant challenge is the low maturity level of stakeholder management practices, particularly within small and medium-sized enterprises (SMEs). Klaus-Rosińska and Iwko [38] highlight that many small construction firms lack well-developed approaches to stakeholder management, including stakeholder identification and analysis, thereby severely impacting project outcomes. This deficiency is surprising given the recognized importance of stakeholder management for achieving sustainable project success. Moreover, the construction industry is characterized by a high degree of stakeholder diversity, which complicates management. This challenge is well documented in literature; for example, Yang et al. [15] note that inadequate stakeholder engagement—exacerbated by a lack of clear objectives and communication—hinders project delivery. The difficulty in identifying and relating to “invisible” stakeholders is particularly problematic, contributing to disconnection and alienation among key participants. Sohu et al. [39] further illustrate this issue in the context of Pakistan, where complex stakeholder relationships lead to high cost and time overruns, underscoring the critical need for effective stakeholder engagement strategies. Another important finding concerns the need for enhanced communication and collaboration throughout the project life cycle. Research by Charan and Vaardini [40] highlights that a systematic approach, including stakeholder mapping and feedback systems, is essential for anticipating and understanding the dynamic requirements of various stakeholders. Furthermore, inadequate collaboration can lead to conflicts and misunderstandings among partners, as noted by Mashali and Eltantawy [41]. Such discord not only jeopardizes project success but can also foster an adversarial culture within the construction ecosystem. Additionally, environmental and institutional factors in developing countries further complicate stakeholder management. The work of Ebekozien et al. [42,43] underscores the necessity for construction projects to engage stakeholders meaningfully to meet development goals sustainably. Furthermore, Ali et al. [43] emphasize that while construction plays a critical role in national economic growth, the industry continues to grapple with challenges, including project delays and stakeholder interests. Effective management practices and institutional support are crucial in overcoming these barriers.

Although these studies provide insights, the literature remains limited in identifying critical factors that hinder effective stakeholder management. Recognizing these factors is vital for developing strategies that enhance project success.

Upon examining the existing literature in depth, it was found that Yang et al. [36] identified the critical success factors for stakeholder management in the Hong Kong context. Yang et al. [36] focused on identifying the critical success factors for effective stakeholder management and proposed a framework based on practices that help projects perform well. Compared with Yang et al. [36], who focused on success factors enabling effective stakeholder management, this study introduces a different and more comprehensive perspective by examining the barriers and structural constraints that hinder stakeholder engagement, particularly in a developing-country setting. The novelty of this research lies in its multi-stage analytical design, combining a systematic literature review (SLR), a national survey, normalized mean value analysis, exploratory factor analysis, and structural equation modeling. This integrated approach not only identifies 69 challenges and condenses them into four core factors, but also quantifies their effect sizes, offering empirical evidence on which obstacles exert the most significant negative influence. By shifting the focus from “what makes stakeholder management succeed” to “why it fails and what structurally blocks it,” this study provides fresh insights that extend and advance Yang’s framework.

Unlike earlier work, this research employs multiple methods to provide a comprehensive view. While it also uses an online questionnaire, its questions are grounded in an SLR, enhancing objectivity. Moreover, it uniquely assesses the importance of factors that hinder effective management, thereby marking a clear departure from previous studies.

3. Methodology

This research adopts a mixed-methods design that integrates both qualitative and quantitative approaches. The qualitative component is based on a systematic literature review (SLR), whereas the quantitative aspect involves statistical analyses. The methodology is structured into four main phases: conducting the literature review, designing a survey instrument based on the review findings, distributing the survey to the selected sample, performing statistical analysis on the responses, and finally, interpreting the outcomes (Figure 1).

3.1. Identifying the Potential Challenges with Systematic Literature Review

An SLR is a methodical research process that involves a detailed analysis of existing studies on a specific topic, using predefined inclusion and exclusion criteria to assess their findings [44]. This approach allows researchers to systematically identify prior work, thoroughly analyze it, and generate valuable insights. SLRs are especially useful in ensuring reliability and validity, as they offer a transparent, unbiased, and reproducible process for data collection and critical evaluation of literature [45].

A review of scholarship in the construction field indicates that, since 2010, a substantial share of published literature reviews have incorporated SLR methodologies. In addition, a notable rise in the adoption of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines has been observed since 2018 [46]. PRISMA offers a structured and transparent protocol for systematically locating, screening, appraising, and synthesizing relevant studies [47]. Consistent with current expectations for methodological rigor, the present investigation applies the PRISMA framework to conduct a systematic review of research on the factors hindering stakeholder management in the construction industry, following a three-phase review process [48].

3.1.1. Stage I: Planning the Review

Establishing a suitable search strategy is a critical component of the initial phase of an SLR, as it defines the breadth and boundaries of the review. Developing such a strategy requires identifying precise keywords and selecting appropriate databases to ensure comprehensive coverage of the most relevant studies [49]. In this study, the literature search was conducted using Web of Science (WOS).

The WoS is widely recognized as a leading resource for literature reviews, owing to its comprehensive global indexing of influential publications [50]. Moreover, WoS employs sophisticated citation-matching algorithms that are generally considered superior to those used by Scopus [51], supporting its designation as the primary database for the present study [52].

The authors developed a search protocol that combined terms drawn from titles, abstracts, and commonly used synonyms in related scholarship. The aim was to construct a search query broad enough to capture all potentially relevant publications while remaining adequately specific to the study’s scope.

The keywords were grouped into three thematic categories:

(1) Stakeholder Management-related terms, such as “stakeholder management” and “stakeholder”;

(2) Sector-related terms, including “construction industry” and “construction sector”;

(3) Terms describing barriers, such as “challenges,” “obstacles,” “barriers,” and “hindrances.”

To initiate the search process, relevant keywords were generated using the “building blocks” technique [53], whereby the research problem is decomposed into key concepts and linked through using Boolean operators such as “AND” and “OR.” Articles were screened based on the presence of these keywords in their titles and abstracts, which served as the principal filtering criteria. These categories were integrated into the following search expression: (“stakeholder management” OR “stakeholder”) AND (“construction industry” OR “construction sector”) AND (“challenges” OR “obstacles” OR “barriers” OR “hindrances”). A structured search was conducted in May 2025. Consistent with the guidance provided by Stekelorum [54], the search omitted books, book chapters, calls for papers, and Special Issue introductions, restricting the dataset to peer-reviewed journal articles to ensure greater rigor and reliability.

Having established the search protocol and selection criteria in this stage, the process proceeded to Stage II, “conducting the review”, during which these parameters were implemented to retrieve, screen, and filter studies in accordance with the PRISMA framework.

3.1.2. Stage II: Conducting the Review

The search for relevant literature was conducted from 2000 to 2024.

Using the protocol described above, the SLR initially yielded 4988 publications. To maintain the quality and relevance of the dataset, explicit inclusion and exclusion criteria were applied. These criteria were designed to ensure that only studies aligned with the research objectives and questions were retained. The inclusion criteria were (1) studies that directly investigate challenges or constraints related to effective stakeholder management within the construction industry and (2) articles published in peer-reviewed scholarly journals. Prioritizing peer-reviewed sources is widely recognized as a means of ensuring rigorous and credible evidence, as highlighted by Shi et al. [55].

The exclusion criteria comprised (1) publications written in languages other than English, (2) studies focused primarily on general stakeholder management and studies related to other industries, and (3) documents for which full-text versions were not accessible. Based on these criteria, eight non-English articles were removed from the dataset. In addition, 156 articles were removed because they were relevant to other industries. Subsequently, the remaining abstracts were screened, and only those that offered substantial discussion of barriers to effective stakeholder management were retained. Papers that mentioned challenges only superficially (i.e., with one to three brief references) were excluded. This step resulted in the removal of 4511 articles, leaving 313 papers for detailed assessment.

In the final phase, full-text evaluations were conducted to determine each paper’s relevance to the research aims, particularly regarding barriers to stakeholder management in the construction industry. This review process produced a final sample of 35 articles. Figure 2 illustrates the complete screening procedure.

To ensure transparency and replicability, the final set of 35 studies included in the SLR is summarized in

Table 1. Summary of the included studies.

Order	Author(s) and the Paper	Year	Country	Sample/Method	Key Findings
1	Olander and Landin [11]	2005	Sweden	2/Case study	They identified that stakeholders influence changes across project phases, and mismanaged stakeholder power/interest leads to appeals, cost overruns, and long-term delays.
2	Yang et al. [15]	2009	Not applicable due to literature review	159/Literature review	Few tools exist to identify all stakeholders; limited research addresses changes in stakeholder influence; and current methods rarely capture the full stakeholder relationship network.
3	Xia et al. [30]	2018	Not applicable due to literature review	79/Content analysis	Four risk–stakeholder linkage modes were determined: (1) management of risk based on stakeholder identification, (2) internal stakeholders’ responsibility and ability in the RM process, (3) management of stakeholder differences concerning risk, and (4) interrelatedness between RM and SM and effect on project performance.
4	Yang et al. [31]	2011	Not applicable due to literature review	68/Content analysis	Identified four major research gaps in construction stakeholder management and addressed them by developing a ranked list of 15 critical success factors.
5	Mok et al. [32]	2015	Not applicable due to literature review	85/Content analysis	Four topics have been identified as the key research themes of Stakeholder management studies in relation to mega construction projects, namely (1) stakeholder interests and influences, (2) stakeholder management process, (3) stakeholder analysis methods, and (4) stakeholder engagement
6	Yang et al. [36]	2010	Hong Kong	183/Quantitative analysis with Spearman’s Rank Correlation	Identified and ranked 15 factors
7	Mashali et al. [37]	2023	Qatar	235/Quantitative analysis with Relative importance index	Communication, early stakeholder engagement, and transparent evaluation are the most influential for improving stakeholder management
8	Ebekozien et al. [42]	2024	South Africa	25/Qualitative analysis with thematic analysis	Three groups ((individual perceived hindrances, organizational perceived hindrances and government-related) perceived hindrances identified.
9	Wang et al. [56]	2022	China	18/Case study with Social Network Analysis (SNA)	Dynamic stakeholder relationship analysis via SNA reveals evolving stakeholder roles and interactions in a Chinese urban regeneration project, offering guidance to enhance stakeholder management.
10	Doloi [57]	2012	Australia	25/Social network analysis	They identified stakeholder influence and integrated stakeholders’ perceptions to evaluate social performance, resulting in a Social Performance Index that highlights the importance of cultural and political stakeholders in project decision-making.
11	Nguyen et al. [58]	2020	Vietnam	157/SEM	They found the project manager leadership competency and project integration management are associated with project management performance.
12	Zhang et al. [59]	2015	China	Axiology-based value analysis	It proposes an axiology-based analytical framework to quantify and integrate diverse stakeholder value perceptions in building projects
13	Al-Fadhali [60]	2024	Yemen	283/SEM	Internal stakeholders such as designers, owners, suppliers, and subcontractors have a significant positive influence on construction project delivery performance.
14	Sanda et al. [61]	2022	Nigeria	61/Quantitative analysis with descriptive statistics and relative index	Conflict of interest and lack of trust among stakeholders are the main drivers of opportunistic behavior in Public–Private Partnership housing projects
15	Karlsen et al. [62]	2008	Norway	Qualitative case study	Trust in project-stakeholder relationships is built through improved communication, reliable and competent behavior, sincerity, integrity, commitment, and shared goals.
16	Lewicki et al. [63]	1998	USA and Singapore	Not applicable due to theoretical rather than empirical	Trust and distrust are distinct but co-existing dimensions in relationships, and develops a multidimensional theoretical framework
17	Atkin and Skitmore [64]	2008	Not applicable due to not being an empirical search	Not applicable due to review and conceptual analysis	Importance of stakeholder management in construction and frames its evolution toward integrating risk, ethics, and sustainability in practice
18	Li et al. [65]	2018	China	22/Delphi Approach	Government bodies are the most influential stakeholders in sustainable construction decision-making in China, with end-users also having high potential influence, and it highlighted the need for more inclusive, transparent participation processes
19	Mok et al. [66]	2017	China	1/Interviews and case analysis	Limited stakeholder involvement, unclear roles, and poor communication are major challenges in major public engineering projects, and that using network-based stakeholder analysis can improve understanding of stakeholder influence
20	Zwikael et al. [67]	2023	Australia	27/Qualitative analysis with interviews	Five project management concepts (project benefits, the iron triangle, critical path, uncertainty, and project leadership) are difficult for stakeholders to understand
21	Bal et al. [68]	2013	UK	Qualitative interviews	The study proposes a six-step stakeholder engagement process and highlights that understanding stakeholder sustainability agendas and measuring their performance are essential to achieving sustainability in construction.
22	Zhao et al. [69]	2015	Singapore	35/Quantitative analysis with Spearman rank correlation	The study found that improved decision-making drives ERM adoption while insufficient resources are the main barrier in Singapore-based Chinese construction firms, with 13 drivers and 22 hindrances identified.
23	Xue et al. [70]	2020	China	108/PLS-SEM	Formal stakeholder relationships dominate cost, quality, safety, and coordination outcomes, whereas informal relationships enhance communication and transparency in Chinese megaproject performance.
24	Mok et al. [71]	2018	Hong Kong	Mixed approach with interviews and surveys.	Stakeholder complexity in Hong Kong green building projects can be analysed holistically using interviews, surveys, and social network analysis, revealing key concerns that influence project success.
25	Zhang et al. [72]	2025	China	Two-mode social network analysis	Collaboration networks among key stakeholders (contractors, designers, manufacturers) are crucial to managing and reducing the high capital cost of prefabricated construction in China through coordinated action on core cost factors.
26	Ebekozien et al. [73]	2023	Nigeria	20/Qualitative analysis with thematic analysis	Nigerian stakeholders know about integrated project delivery, its implementation is limited due to technological, legal, financial, and cultural hindrances.
27	Yang and Zou [74]	2014	China	Theoretical and analytical stakeholder-associated risk analysis model.	It develops a model showing that interconnected stakeholder-associated risks significantly influence green building project outcomes, highlighting the need to manage stakeholder interactions to mitigate risk.
28	Nguyen et al. [75]	2019	Vietnam	Qualitative multi-case study approach	Stakeholders use seven distinct influence strategies, such as withholding inputs, coalition building, and communication, to affect project outcomes in Vietnam
29	Ali and Haapasalo [76]	2023	Finland	Qualitative case study and content analysis	Collaborative stakeholder relationships develop through stages of cooperation, control, and coordination, and that shared understanding and certain preconditions are critical for successful collaboration in Finnish hospital construction.
30	Rafeh et al. [77]	2023	Pakistan	89/Quantitative analysis with factor analysis	18 critical success factors for stakeholder management in Pakistani construction projects were identified, highlighting mission formulation, communication, and stakeholder identification as most important and grouping them into four thematic categories.
31	Safapour et al. [78]	2019	Multiple regions	40/Case studies	Clear project goals, effective stakeholder management, and resource-related communication indicators strongly influence internal communication quality among primary and secondary stakeholders across 40 global construction projects.
32	Yu et al. [79]	2019	China	Bayesian network model	The study shows contractors most strongly influence quality defects, driven by component defects, worker misoperations, and weak inspection.
33	Ekung et al. [80]	2014	Nigeria	186/Mix approach with semi-structured interviews and questionnaire survey	Regulatory gaps, project location, cumulative development effects, poverty, and poor information disclosure are key factors negatively impacting stakeholder engagement outcomes in Nigerian construction projects.
34	Saad et al. [6]	2022	Pakistan	300/SEM	Better stakeholder management leads to greater project success, and this effect is amplified when project teams have higher awareness of stakeholder management practices.
35	El-Sawalhi and Hammad [81]	2015	Gaza Strip	67/Quantitative methods with relative importance index	Effective stakeholder management is driven by project manager competence, transparent evaluation, communication, shared goals, and understanding stakeholder needs.

. The table presents each study’s author(s), publication year, country/region, research method or sample characteristics, and key findings related to stakeholder management barriers.

With the eligible studies summarized in Table 1 identified through the multi-step screening in the existing stage, Stage III, “reporting the review” focused on extracting, consolidating, and reporting the stakeholder management challenges derived from the final pool of publications.

3.1.3. Stage III: Reporting the Review

To develop the list of 69 stakeholder management challenges (SMCs), all 35 articles included in the final SLR stage were examined in full. Each paper was reviewed line by line, and every statement referring to a difficulty, barrier, obstacle, or constraint related to stakeholder management in construction was extracted. During this process, the authors recorded the original wording in each study, along with the surrounding context, to ensure accurate interpretation.

Because different scholars often used varied terms to describe similar problems, an iterative coding and consolidation process was applied. First, all extracted items were grouped based on semantic similarity. Then, challenges that referred to the same underlying issue—such as “poor communication,” “ineffective information exchange,” or “lack of communication channels”—were merged into a single, more representative challenge label: “Ineffective communication between stakeholders.” When two expressions appeared related but not identical, both authors reviewed the texts jointly and agreed on whether to combine them or retain them separately. This constant-comparison technique prevented duplication and ensured that no distinct challenge was lost.

Through this systematic synthesis, 69 unique and clearly defined SMCs were identified. Each challenge in

Table 2. The challenges of effective stakeholder management.

Codes of Challenges (SMCs)	Challenges	References	Source Count (n)
SMC1	Lack of trust among stakeholders	[30,32,56,57,58,59,60,61,62,63,64,65,66]	13
SMC2	Ineffective communication between stakeholders	[11,15,30,31,32,56,57,58,59,60,63,65,66,67,68,69,70,71]	18
SMC3	Poor coordination among project participants	[11,30,31,36,42,56,58,59,61,63,65,66,68,69,71,72]	16
SMC4	Conflicts of interest between key stakeholders	[11,31,32,58,59,61,65,66,73,74,75]	11
SMC5	Disputes between project stakeholders	[11,15,30,31,32,58,59,60,61,63,65,66,68,71,75]	15
SMC6	Lack of commitment by project stakeholders	[32,36,58,61,66,73]	6
SMC7	Lack of experience and competence among contractors and subcontractors	[56,58,59,60,61,62,76]	7
SMC8	Insufficient skills and qualifications of workers	[58,59,61]	3
SMC9	Insufficient professional knowledge and ability among consultants	[56,62,76]	3
SMC10	Poor organizational structure and management competence of project teams	[11,30,36,56,58,59,60,61]	8
SMC11	Inappropriate selection of contractors and subcontractors	[56,62,76]	3
SMC12	Lack of qualified professionals and managers	[60,61,76]	3
SMC13	Ineffective decision-making processes and slow decision-making	[15,30,31,56,57,58,59,60,66,75,77,78]	12
SMC14	Delays in approval and handover processes	[11,57,58,60,66,75,77,78]	8
SMC15	Uncertainty and lack of responsibility in problem-solving	[11,30,56,58,59,60,78]	7
SMC16	Lack of professional responsibility	[56,58,60]	3
SMC17	Delays of owners in solving problems	[56,58,60,67]	4
SMC18	Constant change of subcontractors	[56,60,76]	3
SMC19	Lack of alternative subcontractors	[56,57,60]	3
SMC20	Lack of long-term relationships with suppliers	[57,60,76]	3
SMC21	Bureaucracy in the organization of the owner	[60,62,67,76,79]	5
SMC22	Poor business integration between owner, consultant, and contractor	[60,62,67,76,78]	5
SMC23	Lack of an on-site quality management system	[62,65,76,79]	4
SMC24	Lack of stakeholder participation in the formulation of business policies	[6,11,36,42,66,71,74,80,81]	9
SMC25	Issues with the choice of stakeholder leadership	[6,11,32,36,65,66,68,70,80,81]	10
SMC26	Lack of awareness, interest, and motivation	[11,60,62,66,74,76,80]	7
SMC27	Lack of support and commitment from management and owners	[62,76,80]	3
SMC28	Resistance to change and conservative attitudes	[11,36,60,62,66,76]	6
SMC29	Lack of responsive and ethical behavior	[11,60,62,63,71,80]	6
SMC30	Changes made by the customer and high demands from customers	[58,59,80]	3
SMC31	Privacy policies of project stakeholders	[11,32,58,59,64,66,70,80]	8
SMC32	High demands on the environment, safety, and time by the owner	[58,61,64]	3
SMC33	Transfer of uncertain or unready construction sites to contractors	[58,76,80]	3
SMC34	Stakeholder fatigue and engagement style (active and proactive)	[30,32,36,42,58,66,68,70,71,74,75,76,80]	13
SMC35	Political/social instability	[56,57,58,60,61,66,72]	7
SMC36	New regulations by local governments (taxes, labor, safety, waste, environment)	[11,56,58,61,66,72]	6
SMC37	Bureaucratic and complex procedures	[11,57,58,60,66]	5
SMC38	Excessive approval procedures	[11,58,61,66,72]	5
SMC39	Workers’ strikes	[58,60,61,72]	4
SMC40	Contract disputes and terminations	[56,57,60,61,64]	5
SMC41	Incomplete or unclear contract terms	[57,61,64,72]	4
SMC42	Poor contract management	[56,58,60,64]	4
SMC43	Construction process that violates local laws and regulations	[37,56,61]	3
SMC44	Ecological damage	[37,56,61]	3
SMC 45	Poor data accuracy and unwillingness to save data	[15,37,56,61,70]	5
SMC46	Customers’ impact on data (privacy concerns)	[15,37,56,61]	4
SMC47	Poor data sharing and storage mechanisms	[15,37,56,61,70]	5
SMC48	Vague descriptions of information and documents	[15,56,58,60,70]	5
SMC49	Delays in document approval by consultants or owner-representatives	[58,60,75,78,80]	5
SMC50	The owner’s unclear stance on the project scope, specifications, cost, and schedule	[56,59,61,78]	4
SMC51	Unnecessary variation orders in project design and specifications	[56,59,61,64]	4
SMC52	Different uses and purposes of data by customers and contractors	[37,56,60]	3
SMC53	Low-quality goods and services from suppliers and subcontractors	[56,57,60,76]	4
SMC54	Supply deficiencies in the transportation and delivery of materials, equipment, and labor	[56,57,76]	3
SMC55	Material shortages and inadequate suppliers	[56,60,76]	3
SMC56	Quality defects in building materials	[56,60,61]	3
SMC57	Inefficiencies in material use and resource management	[60,61,76]	3
SMC58	Design changes in the project	[56,57,58,60,61,77,78]	7
SMC59	Inconsistent scale of the project	[57,58,60,77,78]	5
SMC60	Frequent design changes and mistakes made by designers	[56,58,60,61]	4
SMC61	Lack of clear definition of terms such as construction cost	[37,56,59,60,77]	5
SMC62	Local financing and payment problems, lack of available funding	[37,56,60,77]	4
SMC63	Lack of fiscal incentives, regulations, and government support	[57,58,76]	3
SMC64	Lack of technological resources, skills, and knowledge of the supplier or subcontractor	[37,60,77]	3
SMC65	Shortage of technical staff	[37,60,77]	3
SMC66	Language barriers and translation issues	[11,34,57,62]	4
SMC67	Cultural differences and misunderstanding of personal values and beliefs	[11,34,57,62]	4
SMC68	Defective monitoring and control systems for quality, cost, and time	[56,57,61,70]	4
SMC69	Wrong attitudes of workers on quality, cost, environment, safety, etc.	[56,57,61,64]	4

, therefore, reflects a consolidated description that captures the common meaning across multiple sources while eliminating redundancies arising from different terminology used in the literature.

The list of 69 distinct challenges associated with SMCs is presented in Table 2.

3.2. Organizing Questionnaire and Data Collection

The questionnaire was structured into three parts. The first measured participants’ knowledge of stakeholder management. The second assessed 69 SMCs identified through the SLR using a 5-point Likert scale. The final part included eight demographic questions on gender, age, education, profession, organization type, field of work, job position, years of experience, and tenure in the current organization.

Before conducting the main survey, a pilot study was conducted to assess the clarity of the questions, remove ambiguous statements, and estimate the time needed to complete the questionnaire. The draft survey was reviewed by five experts from the construction sector—five architects, five civil engineers, five contractors, and five supplier representatives—each with more than 10 years of professional experience. Their feedback was used to make the necessary revisions and produce the final version of the questionnaire.

The research population comprised architects, civil engineers, contractors, and suppliers, representing key stakeholder groups in Türkiye’s construction sector. A non-probability purposive sampling approach was used to collect data from construction professionals with direct experience in stakeholder management. This method was selected to ensure respondents had the practical knowledge needed to evaluate the identified challenges. The survey link was shared with 20 regional branches of the Turkish Chamber of Architects, 26 branches of the Chamber of Civil Engineers, 30 procurement firms, 21 members of the Contractors Association, and 30 contracting companies, for a total of 500 construction professionals. Participation was voluntary.

Data collection was conducted from 23 August to 22 November 2024, and a total of 170 construction professionals submitted questionnaires. However, 6 were excluded due to missing data, leaving 164 valid responses, for a response rate of 32.8%, which is acceptable for survey-based studies on construction management, where response rates commonly range from 20% to 35% [82]. In the context of construction management studies, determining an acceptable response rate for questionnaires is crucial as it influences the reliability and generalizability of research findings. The literature suggests that response rates of at least 20% to 30% are often considered acceptable for producing credible results in this field [83].

The minimum sample size was calculated using Gamil et al.’s [84] formula (Equation (1)):

$$SS={\displaystyle \frac{Z^{2 }\times P(1-P)}{C^2}}$$

(1)

where

SS = sample size;

Z = z-score (1.96 at a 95% confidence level);

P = the proportion of the population expected to choose an option (0.5 assumed);

C = margin of error (9%).

By applying the formula,

$$SS={\displaystyle \frac{{1.96}^{2 }\times 0.5(1-0.5)}{{0.09}^2}}=118.57\approx 119 (as the minimum sample size)$$

(2)

A 9% margin of error was used in determining the minimum required sample size. This value was selected to balance statistical precision with the practical challenges of obtaining survey responses from construction industry professionals, who often have limited availability [85,86]. While a smaller margin of error (e.g., 5%) would require a substantially larger sample, a 9% margin still provides a reliable representation of perceptions within the target population, especially when combined with robust analytical techniques such as EFA, CFA, and SEM. Sensitivity checks indicated that the primary factor structure and statistical relationships remained stable, suggesting that the chosen margin of error did not materially influence the study’s findings.

Thus, at least 119 responses were required. To further assess sampling adequacy, the marginal error was calculated using the formula outlined by Enshassi and AlSwaity [87]. For a 95% confidence level, the maximum error was $\approx {\displaystyle \frac{1.96}{\sqrt{SS}}}={\displaystyle \frac{1.96}{\sqrt{119}}}=0.18>0.09.$ The margin is considered acceptable, with a minimum size requirement of 119.

Previous studies indicate that for SEM analyses, an adequate sample size typically ranges from 100 to 400 participants [86]. Determining an appropriate sample size is crucial, as it directly affects the reliability of estimated parameters [88]. Iacobucci [89] also notes that, in some cases, sample sizes as small as 50 or as large as 100 may be sufficient. Furthermore, in exploratory factor analysis, researchers often recommend a minimum sample size of 100 participants; however, the literature frequently recommends a sample size of 125–200 to achieve a valid factor structure and maintain statistical power [90]. In the context of this study, a sample size of 164 is considered modest but adequate.

Table 3. The percentage and frequency distributions of the sample group’s demographic characteristics.

Demographic Variables		Frequency (f)	Percent (%)
Gender	Female	56	34.1
	Male	108	65.9
Age	20–30	64	39.0
	31–38	56	34.1
	39–46	24	14.6
	47–54	12	7.3
	55 and over	8	4.9
Educational Background	Primary/Secondary Education	14	8.5
	High school	25	15.2
	Undergraduate/University	95	57.9
	Master	28	17.1
	Ph.D.	2	1.2
Profession	Architect	61	37.2
	Civil Engineer	44	26.8
	Contractor	29	17.7
	Supplier	30	18.3
Type of company	Public Institution	12	7.3
	Private Company	152	92.7
Work Site	Office	34	20.7
	Site	13	7.9
	Office + Construction Site	117	71.3
Position in the company	Administrator	109	66.5
	Employee	55	33.5
Working time in the construction industry	1–5 years	51	31.1
	6–10 years	52	31.7
	11–15 years	25	15.2
	16–20 years	16	9.8
	21 years and older	20	12.2

presents the demographic distribution of the 164 valid participants included in the analysis.

The demographic profile of the respondents aligns with the structure of Türkiye’s construction industry. National labor statistics indicate that the sector remains heavily male-dominated, which explains the gender imbalance observed in the dataset. Likewise, most participants were employed by private firms, reflecting the private sector’s prominent role in project delivery across the country. These characteristics do not undermine the study but rather mirror the current workforce composition. Nevertheless, the findings should be interpreted with this context in mind, and future studies involving more balanced and diverse samples across multiple regions would help enhance the generalizability of the results.

3.3. Data Analysis

The quantitative analysis followed a structured, step-by-step workflow to ensure that each technique built logically on the previous stage. First, descriptive statistics and reliability tests were conducted to confirm that the 69 SMCs were internally consistent and suitable for further analysis. The survey instrument’s reliability was assessed using Cronbach’s alpha, with values above 0.70 indicating acceptable internal consistency [91].

Next, descriptive statistical analyses were conducted. Testing data normality is essential in quantitative research; thus, skewness and kurtosis were calculated. Based on this, mean and standard deviation values were computed for the 69 SMCs.

After this validation, the NMV analysis was applied to identify which challenges respondents perceived as most influential. NMV reduced the complete list to a smaller set of critical challenges, thereby focusing subsequent analyses on the items with the highest practical importance. An NMV analysis was then applied to identify the most critical challenges (Equation (3)):

$$Normalized mean value={\displaystyle \frac{(mean of SMC - lowest ranked mean)}{(highest ranked mean - lowest ranked mean)}}$$

(3)

The NMV for each SMC was computed using Equation (3). SMCs with an NMV greater than 0.5 were classified as critical challenges (CCs), consistent with the approach adopted by Xu et al. [92], Zhao et al. [69], and Liao and Teo [29]. To reinforce identification of these CCs, the average score for each SMC was also compared with the overall mean of all SMCs. If an SMC’s mean value was higher than the overall mean, it was likewise categorized as a critical challenge. This supplementary assessment technique was previously applied by Won et al. [29,93] and again by Liao and Teo [29].

Once the critical challenges were identified, an EFA was performed to uncover the underlying structure among these items. EFA grouped critical challenges into broader latent dimensions, termed critical stakeholder management factors (CSMFs), thereby reducing data complexity and revealing how individual challenges clustered conceptually.

EFA involves four main stages: data preparation, factor extraction, rotation, and interpretation. Dataset suitability is typically tested using the Kaiser–Meyer–Olkin (KMO) measure and Bartlett’s Test of Sphericity. A KMO above 0.5 and a significant Bartlett’s Test confirm appropriateness [94]. Once suitability is confirmed, factor extraction methods such as Principal Component Analysis (PCA) are applied to determine the optimal number of factors [95].

In this study, the CSMFs identified through EFA were further examined using CFA in LISREL to test their validity. Validity refers to the extent to which a test accurately measures the construct it intends to capture. Survey items with higher validity are more effective in reflecting the targeted characteristics. Model adequacy was evaluated through several indices, including the comparative fit index (CFI), the root mean square error of approximation (RMSEA), and the chi-square (χ²) statistic. Within CFA, path coefficients represent the strength of associations among variables; coefficients below 0.1 denote weak effects, those near 0.3 indicate moderate effects, and values of 0.5 or above suggest strong influences [96]. At the 99% confidence level, coefficients of at least 0.5, combined with t-values above 2.58, were considered statistically significant.

In the final stage, SEM was applied to measure the effect sizes of CSMFs that hinder effective stakeholder management. SEM was preferred because it can represent latent variables, offering a more precise assessment of CSMFs than conventional multivariate regression. Unlike regression analysis, SEM simultaneously estimates both measurement and structural models, providing a broader view of the factors hindering stakeholder management. It also provides model fit evaluation using indices such as CFI, RMSEA, and goodness-of-fit index (GFI), thereby strengthening the robustness of the findings. Beyond this, SEM captures complex interdependencies among multiple dependent and independent variables, enabling the quantification of each CSMF’s relative effect. It further examines how measurement paths align with latent constructs. While there is debate regarding the threshold for acceptable path coefficients, a minimum of 0.2 is commonly recommended [97]. For this study, at the 99% confidence level, path coefficients of 0.5 or greater and t-values exceeding 2.58 were considered statistically significant [98]. Additionally, SEM can address measurement errors and multicollinearity, thereby enhancing the reliability and validity of the outcomes, making it the most appropriate method for this analysis.

Together, these sequential steps—NMV → EFA → CFA → SEM—formed an integrated analysis workflow that progressively narrowed, structured, validated, and modeled the critical challenges identified in the study.

4. Results

4.1. Reliability Analysis

In this study, a reliability test was conducted to assess the internal consistency of 69 perception-based SMCs, measured on a 5-point Likert scale. Cronbach’s alpha (α) was employed as the reliability coefficient, and the analysis produced a value of 0.979. Since values above 0.90 are considered to demonstrate excellent internal consistency, the results confirm that the dataset is highly reliable [99].

4.2. Normality Analysis

The survey items were measured using five-point Likert scales, which are technically ordinal. Therefore, the analysis does not assume strict normality. Skewness and kurtosis were examined to detect severe deviations from symmetry that might affect estimation, rather than to claim that the data were normally distributed. In line with common practice in SEM applications in social science and construction management research, the Likert items were treated as approximately continuous because no extreme non-normality was observed. This enabled maximum likelihood-based estimation while maintaining acceptable model robustness. Values ranging between −3 and +3 are generally considered acceptable indicators of normality [100]. As shown in

Table 4. Definition and ordering of SMCs (n = 164).

Codes of Challenges	Skewness	Kurtosis	Mean	Standard Deviation	NMV	Rank
SMC1	−0.636	0.148	3.80	0.978	0.730 *	28
SMC2	−0.715	0.454	3.88	0.923	0.802 *	13
SMC3	−0.775	0.242	3.94	0.970	0.856 *	9
SMC4	−0.673	−0.075	3.85	1.026	0.775 *	18
SMC5	−0.551	−0.375	3.83	1.025	0.757 *	22
SMC6	−0.841	0.209	3.86	1.042	0.784 *	15
SMC7	−1.283	1.386	4.10	1.025	1.000 *	2
SMC8	−1.105	0.742	4.10	1.038	1.000 *	3
SMC9	−1.085	0.915	4.03	1.009	0.937 *	5
SMC10	−0.989	0.549	3.99	1.033	0.901 *	7
SMC11	−1.104	1.078	4.10	0.967	1.000 *	1
SMC12	−0.915	0.393	4.03	0.956	0.937 *	4
SMC13	−0.523	−0.344	3.76	1.040	0.694 *	31
SMC14	−0.435	−0.430	3.73	1.011	0.667 *	36
SMC15	−0.761	−0.154	3.94	1.038	0.856 *	10
SMC16	−1.096	0.604	4.03	1.059	0.937 *	6
SMC17	−0.961	0.470	3.96	1.015	0.874 *	8
SMC18	−0.912	0.126	3.86	1.145	0.784 *	17
SMC19	−0.812	0.166	3.81	1.094	0.523 *	52
SMC20	−0.491	−0.421	3.57	1.091	0.604 *	49
SMC21	−0.342	−0.604	3.66	1.032	0.766 *	20
SMC22	−0.800	0.140	3.84	1.083	0.685 *	33
SMC23	−0.894	0.359	3.75	1.099	0.649 *	41
SMC24	−0.717	0.062	3.71	1.039	0.459	56
SMC25	−0.503	−0.418	3.50	1.108	0.459	58
SMC26	−0.574	−0.507	3.50	1.157	0.369	61
SMC27	−0.471	−0.348	3.40	1.094	0.477	54
SMC28	−0.582	−0.180	3.52	1.056	0.631 *	44
SMC29	−0.693	−0.231	3.69	1.141	0.766 *	21
SMC30	−0.725	−0.046	3.84	1.033	0.297	64
SMC31	−0.296	−0.621	3.32	1.139	0.459	59
SMC32	−0.551	−0.313	3.50	1.141	0.667 *	37
SMC33	−0.741	−0.174	3.73	1.146	0.667 *	38
SMC34	−0.454	−0.363	3.40	1.083	0.369	60
SMC35	−0.365	−0.671	3.39	1.186	0.360	62
SMC36	−0.693	−0.411	3.75	1.171	0.685 *	34
SMC37	−0.848	−0.206	3.76	1.222	0.694 *	32
SMC38	−0.908	0.197	3.90	1.103	0.820 *	11
SMC39	−0.102	−1.006	3.12	1.255	0.117	67
SMC40	−0.680	−0.265	3.71	1.132	0.649 *	42
SMC41	−0.824	0.140	3.85	1.069	0.775 *	19
SMC42	−0.829	0.272	3.81	1.063	0.739 *	26
SMC43	−0.753	−0.083	3.78	1.123	0.712 *	29
SMC44	−0.273	−0.743	3.28	1.165	0.261	65
SMC45	−0.450	−0.427	3.58	1.048	0.532 *	51
SMC46	−0.340	−0.759	3.35	1.173	0.324	63
SMC47	−0.322	−0.528	3.26	1.116	0.243	66
SMC48	−0.563	−0.234	3.50	1.096	0.459	57
SMC49	−0.732	0.243	3.66	1.061	0.604 *	48
SMC50	−0.659	−0.020	3.69	1.020	0.631 *	43
SMC51	−0.775	0.120	3.69	1.107	0.631 *	45
SMC52	−0.406	−0.484	3.55	1.095	0.505 *	53
SMC53	−0.978	0.529	3.82	1.088	0.748 *	24
SMC54	−0.623	−0.242	3.73	1.083	0.667 *	39
SMC55	−0.692	−0.222	3.86	1.059	0.784 *	16
SMC56	−0.785	0.078	3.81	1.086	0.739 *	27
SMC57	−0.892	0.409	3.89	1.024	0.811 *	12
SMC58	−0.810	0.071	3.77	1.103	0.703 *	30
SMC59	−0.502	−0.532	3.73	1.075	0.667 *	40
SMC60	−0.719	0.176	3.81	1.016	0.739 *	25
SMC61	−0.664	−0.326	3.69	1.136	0.631 *	46
SMC62	−0.853	−0.163	3.83	1.167	0.757 *	23
SMC63	−0.708	−0.447	3.65	1.245	0.595 *	50
SMC64	−0.903	0.214	3.74	1.148	0.676 *	35
SMC65	−0.940	0.381	3.87	1.095	0.793 *	14
SMC66	−0.004	−0.968	3.01	1.237	0.018	68
SMC67	0.012	−1.032	2.99	1.255	0.000	69
SMC68	−0.401	−0.551	3.52	1.119	0.477	55
SMC69	−0.386	−0.567	3.69	1.060	0.631 *	47

* Refers to critical challenges (CCs).

, the skewness and kurtosis values for all variables fell within this range, confirming that the dataset meets the assumption of a normal distribution.

4.3. Defining and Ordering the Critical Stakeholder Management Challenges

The means, standard deviations, and normalized mean values of 69 SMCs are represented in Table 4.

The NMV analysis revealed that 16 of the 69 identified SMCs (SMC25, SMC26, SMC27, SMC28, SMC31, SMC32, SMC34, SMC35, SMC39, SMC44, SMC46, SMC47, SMC48, SMC66, SMC67, and SMC68) scored below the 0.5 threshold and were therefore excluded from further consideration. In the NMV analysis, SMCs with NMV < 0.5 were excluded because this threshold indicates that respondents rated these items as less influential than the average challenge across the dataset. The 16 excluded SMCs received consistently low scores, meaning that practitioners do not experience these issues as frequent or impactful barriers in practice. Prior studies using NMV similarly apply the 0.5 threshold to filter out low-priority items before conducting factor analysis [28,29]. Removing low-priority items before EFA is recommended [28] to reduce statistical noise, improve factor clarity, and ensure that only challenges perceived as substantively relevant by the industry are used in the subsequent analyses.

Because these challenges are not consistently experienced across projects, they were assigned lower NMVs. Excluding them prior to EFA improved the clarity of the factor structure by reducing noise and allowing the analysis to focus on the most widely recognized and impactful barriers.

The remaining 53 SMCs, with NMVs exceeding 0.5, were classified as critical challenges (Table 4). When two or more SMCs had the same normalized mean score, ranking was determined by the lower standard deviation (SD), as a smaller SD indicates less variability in responses and a more stable mean [101]. Among the results, SMC7, SMC8, and SMC11 achieved the maximum NMV score of 1.000. The most critical challenge was identified as the inappropriate selection of contractors and subcontractors (SMC11) with the smallest SD within the three SMC. Closely following were the lack of experience and competence among contractors and subcontractors (SMC7) and the insufficient skills and qualifications of workers (SMC8), both of which represent major challenges to effective stakeholder management.

4.4. Unveiling the Critical Factors with Exploratory Factor Analysis

An EFA was conducted to identify the underlying structures of the 53 CCs and determine the CSMFs in the Turkish construction industry (

Table 5. The results of exploratory and confirmatory factor analysis.

Factors	Code of SMCs	EFA			CFA
		Eigenvalue	Loads of SMCs	% of Variance	Standardized Coefficients
Factor 1	SMC58	10.025	0.697	18.915	0.76
	SMC61		0.683		0.71
	SMC59		0.672		0.76
	SMC60		0.666		0.78
	SMC54		0.653		0.79
	SMC64		0.634		0.76
	SMC49		0.614		0.73
	SMC57		0.605		0.78
	SMC55		0.588		0.71
	SMC62		0.576		0.71
	SMC52		0.574		0.73
	SMC51		0.560		0.74
	SMC43		0.555		0.74
	SMC40		0.555		0.69
	SMC53		0.555		0.74
	SMC42		0.551		0.76
	SMC45		0.533		0.66
	SMC63		0.530		0.64
	SMC50		0.526		0.73
	SMC56		0.521		0.70
	SMC41		0.518		0.74
	SMC33		0.515		0.67
	SMC69		0.479		0.71
	SMC65		0.468		0.64
Factor 2	SMC19	8.317	0.751	15.692	0.70
	SMC20		0.656		0.59
	SMC18		0.653		0.70
	SMC13		0.638		0.76
	SMC17		0.607		0.71
	SMC15		0.605		0.79
	SMC16		0.587		0.71
	SMC23		0.576		0.78
	SMC11		0.569		0.75
	SMC12		0.567		0.73
	SMC9		0.530		0.64
	SMC21		0.527		0.64
	SMC22		0.494		0.70
	SMC6		0.489		0.72
	SMC7		0.485		0.73
Factor 3	SMC38	6.728	0.719	12.694	0.73
	SMC37		0.713		0.69
	SMC28		0.649		0.65
	SMC36		0.601		0.68
	SMC14		0.585		0.76
	SMC32		0.541		0.70
	SMC29		0.541		0.67
	SMC8		0.430		0.69
Factor 4	SMC2	6.146	0.776	11.596	0.89
	SMC1		0.752		0.85
	SMC3		0.717		0.80
	SMC4		0.650		0.78
	SMC5		0.551		0.73
	SMC10		0.521		0.73
Total Explained Variance				58.897	χ2/df	2.122
Kaiser-Meyer-Olkin (KMO) value				0.917	RMSEA	0.043
Bartlett’s test of sphericity	Approx. Chi-square			6865.468	CFI	0.970
	degree of freedom (df)			1378	GFI	0.910
	p			<0.001

).

Before the EFA, dataset suitability was assessed using the KMO measure and Bartlett’s Test of Sphericity. The KMO value was 0.917, well above the 0.5 threshold, confirming sampling adequacy [101]. Bartlett’s Test was significant (p < 0.001), showing the correlation matrix differed from an identity matrix and was suitable for factor extraction.

The PCA was then applied, retaining components with eigenvalues greater than 1 to ensure each factor explained more variance than a single variable. The analysis produced four factors, accounting for 58.897% of the total variance, which were categorized as CSMFs. In social, behavioral, and management sciences, where constructs are often latent, multi-dimensional, and subject to high measurement error (e.g., perception-based survey data), the standard requirement is typically lower than in the physical sciences. This leads methodologists to suggest that an explained cumulative variance of 50% to 60% is generally considered robust and acceptable. Specifically, Field [102] acknowledges that in many instances of psychological and social data, 50% is sufficient due to the inherent complexity and irreducible measurement error associated with these constructs.

Therefore, an explained variance of 58.897%, as observed in this study, is statistically strong and falls well within the academically recognized benchmark for research involving human perceptions and complex organizational dynamics.

To improve interpretability, Varimax rotation was used, as it maximizes variance among loadings and clarifies groupings. Factor loadings above 0.40 were considered significant, and all critical challenges exceeded this threshold (Table 5). The four extracted CSMFs and their corresponding labels are as follows:

Factor 1: Weak Planning, Coordination, and Implementation Deficiencies (WPCID);

Factor 2: Institutional and Operational Weaknesses (IOW);

Factor 3: Communication Problems (CP);

Factor 4: Legal Regulations, Bureaucratic Barriers, and Ethical Issues (LRBE).

4.5. Confirmatory Factor Analysis (CFA)

Table 5 presents the CFA outcomes, indicating that all factor loadings exceed the 0.5 threshold. The model achieved a χ²/df ratio of 2.122, a comparative fit index (CFI) of 0.97, a root mean square error of approximation (RMSEA) of 0.043, and a goodness-of-fit index (GFI) of 0.91, all of which meet the recommended model-fit criteria. Taken together, these statistics confirm that the CFA model demonstrates strong fit and is suitable for evaluating the reliability of the measurement scales.

Building on the satisfactory model-fit results reported above,

Table 6. The results of the convergent and discriminant validity of CFA.

Factors	CR	AVE
WPCID	0.98	0.56
IOW	0.93	0.51
CP	0.88	0.50
LRBE	0.91	0.64

presents the composite reliability (CR) and average variance extracted (AVE) for each latent variable to assess convergent and discriminant validity further. All CR values range between 0.88 and 0.98, with a recommended threshold of 0.70, indicating strong internal consistency across the measurement items [103]. The AVE values range from 0.50 to 0.64, meeting the minimum criterion of 0.50 suggested by Fornell and Larcker [104]. Overall, the CR and AVE values in Table 6 demonstrate that the measurement model is both reliable and conceptually distinct, providing a strong foundation for subsequent SEM analysis.

4.6. Evaluating the Measurement Model

SEM was employed to examine how the four CSMFs affect stakeholder management. Figure 3 depicts the proposed structural framework, illustrating the hypothesized causal pathways supported by the CFA and reliability analyses. After establishing the validity of the measurement model through CFA, a conceptual SEM was formulated, generating four hypotheses (H1–H4). The model incorporates four latent constructs, each reflecting a key contributor to stakeholder management, and specifies their direct effects on the dependent variable, “Effective stakeholder management.” Each path in Figure 3 represents a theoretical link between two constructs and corresponds to one of the stated hypotheses.

Considering the four latent factors as critical stakeholder management factors, hypotheses were formulated as follows, and paths are shown in Figure 3:

H1: 

WPCID impacts on stakeholder management.

H2: 

IOW impacts on stakeholder management.

H3: 

CP impacts on stakeholder management.

H4: 

LRBE impacts on stakeholder management.

4.7. Reliability and Validity of the Model

To assess the consistency and validity of the latent constructs, Cronbach’s alpha (CA) (α) and composite reliability (CR) were calculated. All constructs achieved CR values between 0.96 and 0.88—well above the recommended benchmark of 0.70. Cronbach’s alpha values ranged from 0.88 to 0.96, likewise exceeding the minimum acceptable level. Internal consistency was further confirmed using the average variance extracted (AVE), with permissible values equal to or above 0.50 [104]. As summarized in

Table 7. The results of reliability and validity tests.

Latent Variables	CR	CA	AVE
WPCID	0.96	0.96	0.53
IOW	0.94	0.94	0.51
CP	0.88	0.88	0.50
LRBE	0.91	0.91	0.64

, AVE values ranged from 0.50 to 0.64, supporting the adequacy of all components. Building on these results, a final SEM was estimated (Figure 3), and its goodness-of-fit indices and t-statistics—presented in Table 7 and

Table 8. Model statistics.

Fit Index	Suggested Values	Structural Equation Results	Evaluation
X2/df	0 ≤ X2/df ≤ 3	2.09	Good Fit
X2	-	2853.55	-
GFI	0. 95 ≤ GFI ≤ 1.00	0.95	Good Fit
AGFI	0. 95 ≤ AGFI ≤ 1.00	0.96	Good Fit
RMSEA	0 ≤ RMSEA ≤ 0.05	0.04	Good Fit
RMSEA 90% CI	<0.06	0.038–0.052	Good Fit
CFI	0. 95 ≤ CFI ≤ 1.00	0.97	Good Fit
TLI	>0.90	0.96	Good Fit
SRMR	<0.08	0.04	Good Fit
NFI	0. 95 ≤ NFI ≤ 1.00	0.95	Good Fit

—validated the robustness of the structural relationships.

Building on the framework in Figure 3, the finalized SEM is presented in Figure 4, while the validation results—comprising GOF indices and t-values—are reported in Table 8 and

Table 9. Estimates of the structural equation model’s path coefficients.

Hypothetical Paths and Expected Influences	Path Coefficient *	t-Value (1-Tail)	Interpretation	R2
H1: WPCID → STKM	−0.89	−10.18	Supported	0.80
H2: IOW → STKM	−0.95	−9.67	Supported	0.90
H3: CP → STKM	−0.92	−9.77	Supported	0.86
H4: LRBE → STKM	−0.81	−10.73	Supported	0.66

Note: * All standardized path coefficient estimates are expected to be significant at p < 0.01.

.

As shown in Table 8 the finalized SEM achieves an acceptable level of fit, meeting the standard GOF benchmarks. The proposed pathways, along with their R² values and effect sizes, are reported in Table 9.

At the 0.01 significance level, all hypotheses surpassed the critical one-tailed t-value of 2.58 (Table 9). The results further reveal that IOW (−0.95) and CP (−0.92) exert the most significant influence on stakeholder management, followed by WPCID (−0.89) and LRBE (−0.81).

The standardized path coefficients in Table 9 are relatively high and negative, ranging from −0.81 to −0.95. These values reflect the strong structural relationships between the four CSMFs and the dependent variable, effective stakeholder management.

First, the measurement model demonstrates excellent reliability and validity, with CR values ranging from 0.88 to 0.96 and AVE values ranging from 0.50 to 0.64, confirming that each construct is measured consistently and without excessive error. The strong internal consistency reduces noise, producing more precise parameter estimates.

Second, all model-fit indices fall within the recommended thresholds (e.g., χ²/df = 2.09, CFI = 0.97, RMSEA = 0.04), indicating that the SEM provides a good representation of the observed data and is not overfitted to sample-specific patterns.

Third, the negative coefficients are conceptually consistent with how the constructs were defined: each CSMF represents a challenge, meaning higher scores indicate more severe barriers. Therefore, a more substantial presence of these barriers naturally results in lower stakeholder management effectiveness, producing stable, strongly negative coefficients.

Finally, the high magnitudes are supported by the large explained variance (R² = 0.66–0.90), showing that the factors collectively account for a substantial proportion of the variance in stakeholder management. This reflects the central role of structural deficiencies, coordination problems, information weaknesses, and regulatory/ethical hurdles in shaping performance.

5. Discussion

Stakeholder management is vital in construction projects due to complexity, diverse interests, and high risks. Multiple stakeholders with varying expectations make effective management essential for aligning interests, mitigating conflicts, and fostering collaboration. Identifying challenges improves efficiency, enhances performance, and supports construction’s social and economic sustainability.

In this study, SEM supported all four hypotheses. Figure 4 and Table 9 illustrate noticeable variations in the path coefficient values across the proposed hypotheses. Coefficients close to 1 typically reflect a very strong linkage, while those approaching 0 suggest a weak connection [103]. Based on these values, the influence of the key factors limiting stakeholder management can be grouped into two categories: coefficients between 1 and 0.81 indicate the most influential group, with a very strong impact, whereas coefficients ranging from 0.60 to 0.40 indicate a more moderate influence. In this study, all path coefficients were 0.81 or higher. The four critical stakeholder management factors are revealed below.

5.1. Institutional and Operational Weaknesses

Institutional and operational shortcomings are another key factor, stemming from structural issues, weak management, and low productivity [8]. Inadequate institutional capacity leads to stakeholder incompatibilities and delays [105], while poor governance limits long-term strategies and sustainable growth [106]. In developing contexts, these weaknesses intensify coordination problems [107].

According to this study’s findings, ICW is the most significant factor hindering stakeholder management, with a path coefficient of −0.95, which aligns closely with longstanding structural issues in Türkiye’s construction industry. The sector is characterized by uneven institutional capacity, frequent organizational restructuring, and varying levels of professional competence across firms [108]. In many projects—especially public or large-scale infrastructure works—bureaucratic fragmentation and overlapping responsibilities between agencies lead to delays in decisions, unclear authority, and inconsistent oversight. At the company level, operational systems such as quality management, documentation practices, and internal communication procedures are often not standardized, particularly among medium-sized contractors and subcontractors who dominate the market. These institutional gaps create uncertainty, slow responses to emerging problems, and hinder coordinated action among stakeholders. Because such weaknesses permeate both the regulatory environment and day-to-day project operations, practitioners in Türkiye experience them as the most influential barrier to effective stakeholder management.

5.2. Communication Problems

Stakeholder management is vital in construction projects, yet the large number of actors complicates coordination and often hinders communication [12]. CP has emerged as the second most influential factor, with a −0.92 path coefficient, which is consistent with common dynamics in Türkiye’s construction sector. Projects frequently involve multiple contractors, subcontractors, and suppliers working simultaneously, yet communication channels among these parties are often informal or fragmented [109]. Many firms—especially small and medium-sized ones—still rely on verbal instructions, informal communication tools such as WhatsApp messages, or undocumented agreements instead of standardized information systems. This creates inconsistencies in the transfer of design updates, site decisions, and stakeholder expectations. In large public or infrastructure projects, communication is further complicated by hierarchical structures and long reporting chains, which slow information flow and increase the likelihood of misunderstandings. Additionally, rapid project timelines and frequent scope changes can make maintaining timely, transparent communication difficult. These conditions explain why practitioners view communication-related issues as a significant barrier: they directly contribute to coordination failures, disputes, and misaligned expectations among stakeholders in Türkiye’s construction environment.

5.3. Weak Planning, Coordination, and Implementation Deficiencies

Weaknesses in planning and coordination affect all project stages, causing delays, cost overruns, and quality issues [107,110]. These problems are especially severe in developing countries with limited resources and flawed time management. Poor project management erodes trust, fuels conflicts, and increases the risk of failure [57,111]. Coordination problems also trigger design errors and rework, compounding delays [112].

Based on this study’s results, WPCID is the third most influential critical factor hindering stakeholder management, with a path coefficient of −0.89. Although this factor ranks third in terms of influence, its importance is closely linked to the structural features of Türkiye’s construction industry. Many projects operate under accelerated timelines driven by competitive tendering, political visibility, and rapid urban development pressures [113]. These conditions often reduce the time allocated for detailed planning and limit opportunities for early stakeholder engagement. In addition, the sector relies heavily on multilayered subcontracting, which can disrupt workflow coordination and create gaps in responsibility during implementation. Public sector projects frequently involve complex approval steps and mid-project scope adjustments, further weakening coordination efforts. As a result, even though other factors exert a more substantial overall influence, deficiencies in planning, coordination, and implementation remain a significant barrier in Türkiye because they amplify misunderstandings between stakeholders and contribute to avoidable delays and rework.

Effective planning and coordination are, therefore, essential. In large-scale projects, modern approaches such as Integrated Project Delivery (IPD) help mitigate these challenges [114].

5.4. Legal Regulations, Bureaucratic Barriers, and Ethical Issues

Another significant challenge in construction stakeholder management involves complex, frequently changing legal regulations, bureaucratic hurdles, and ethical concerns. LRBE is the fourth significant factor, yet it remains a notable challenge due to the regulatory landscape of Türkiye’s construction industry. Construction projects often require approvals from multiple public agencies, each with its own procedures, timelines, and documentation standards. This creates lengthy administrative cycles and uncertainty for project teams. Frequent changes to regulations—especially those related to zoning, safety, environmental requirements, and procurement—can also disrupt planning and require sudden adjustments [113]. In addition, issues such as inconsistent enforcement, varying interpretations of regulations across municipalities, and occasional ethical concerns in tendering or inspection processes further complicate stakeholder interactions. These systemic inefficiencies slow decision-making, limit transparency, and hinder trust between project participants. As a result, while this factor ranks lower than the others, it remains a meaningful constraint in Türkiye’s construction environment, influencing how smoothly stakeholders can collaborate and comply with project requirements.

6. Conclusions

This study employed a comprehensive methodological framework to identify the critical factors hindering effective stakeholder management. An SLR yielded 69 SMCs, which informed the development of a questionnaire. Data collected from 164 respondents were analyzed using the NMV approach, which highlighted 53 SMCs as critical challenges. Subsequently, EFA grouped these into four overarching CSMFs, namely, weak planning, coordination, and implementation deficiencies (WPCID); institutional and operational weaknesses (IOW); communication problems (CP); and legal regulations, bureaucratic barriers, and ethical issues (LRBE).

SEM was utilized to examine the effects of these CSMFs. The results indicated that the most influential CSMFs were IOW and CP, with path coefficients of −0.95 and −0.92, respectively. Additionally, WPCID and LRBE emerged as significant contributors, with path coefficients of −0.89 and −0.81, respectively.

Based on the SEM findings, the four factors were ranked by their impact on stakeholder management performance. Institutional and operational weaknesses emerged as the most influential factor, followed by communication problems, planning/coordination deficiencies, and finally legal and bureaucratic/ethical issues. This ranking guided the prioritization of intervention measures.

6.1. Practical, Conceptual, and Empirical Implications

This study identifies four critical factors in Türkiye’s construction industry—weak planning and coordination; institutional and operational weaknesses; communication problems; and legal, bureaucratic, and ethical challenges—with implications for practitioners, policymakers, and researchers. Addressing them requires practical interventions, institutional reforms, and theoretical advances.

Institutional and operational weaknesses in the construction sector: implications for practitioners and policymakers

Given its significant negative impact on stakeholder management, improving institutional and operational mechanisms should be the priority. Practitioners may establish stakeholder-management protocols (stakeholder mapping, engagement plans, responsibility charts) to ensure that all critical stakeholders are identified early and integrated systematically throughout the project lifecycle. Furthermore, practitioners may create cross-departmental stakeholder coordination units or assign stakeholder coordinators to align design, site, and management teams with stakeholder expectations. On the other hand, policymakers should mandate standardized stakeholder management frameworks (e.g., templates for stakeholder analysis, engagement logs, and conflict-resolution protocols) to ensure consistency across firms. In addition, they should introduce certification or compliance requirements for organizational systems that enhance stakeholder coordination and transparency.

Communication problems in construction projects: implications for practitioners and policymakers

Since communication problems are the second most influential factor, construction organizations should address them. Within this scope, practitioners may implement structured communication pathways for all stakeholder interactions, including formal procedures for reporting design changes, documenting client decisions, and sharing updates with contractors/subcontractors. Additionally, practitioners may use digital platforms that track stakeholder actions—such as issue-tracking tools, approval workflows, and real-time collaboration environments—to ensure no stakeholder input is lost or misinterpreted. Developing minimum communication standards that require firms to maintain traceable stakeholder communication records, including meeting minutes, change requests, and formal approvals. Additionally, policymakers may encourage or require the adoption of common data environments (CDEs) to support transparent, multi-stakeholder information sharing.

Weak planning, coordination, and implementation deficiencies in the construction process: implications for practitioners and policymakers

To minimize deficiencies in planning, coordination, and implementation in the construction process, practitioners may integrate stakeholders into early planning stages through co-design workshops, risk-planning sessions, and joint scheduling meetings to avoid downstream misunderstandings. Furthermore, practitioners use BIM-supported stakeholder coordination models (e.g., 4D simulations, design reviews) to visualize impacts of stakeholder decisions and reduce interface conflicts. On the other hand, policymakers should require stakeholder-inclusive planning guidelines that define how public agencies, clients, contractors, and communities must be consulted and engaged during primary project stages. Additionally, they should develop evaluation criteria that assess how well firms incorporate stakeholder feedback into planning, risk assessment, and coordination processes.

Legal regulations, bureaucratic barriers, and ethical issues in the construction sector: implications for practitioners and policymakers

Although ranked fourth, legal regulations, bureaucratic barriers, and ethical issues require attention due to their systemic impact. To minimize this issue, practitioners should use standardized and transparent contract clauses that clearly define stakeholder roles, communication responsibilities, escalation procedures, and dispute-resolution mechanisms. Furthermore, they may implement ethical stakeholder-engagement policies—including transparent reporting, anti-bribery measures, and fair consultation practices—to reduce conflict and maintain trust. When the issue is considered within the scope of policymakers, they should streamline approval processes involving multiple stakeholders (municipalities, utilities, regulatory bodies) to reduce delays and improve coordination. Finally, policymakers should develop legal guidelines for stakeholder engagement that clarify authority, establish documentation standards, and specify expected timelines for stakeholder approvals.

6.2. Educational Implications

This study’s findings have several implications for construction education and professional development. Critical barriers such as inappropriate contractor selection, workforce qualification gaps, and subcontractor competence highlight the need to integrate stakeholder management into architecture, engineering, and construction (AEC) curricula. Courses on procurement, contract administration, conflict resolution, and collaborative project delivery, along with case-based learning and simulations, can prepare students for complex stakeholder environments.

Continuous professional development programs should target contractors, subcontractors, and consultants, focusing on enhancing their communication, negotiation, and collaboration skills. Stronger university–industry partnerships, where students engage in live projects, can bridge theory and practice. Curricula should also cover construction law, ethics, and governance to equip graduates with the policy literacy and ethical decision-making skills necessary for informed policy-making. While based in Türkiye, these insights are relevant to other developing economies with similar challenges.

6.3. Limitations and Future Research Directions

This study has some limitations. The SLR used only the WoS database; including Scopus or Google Scholar in future reviews could broaden coverage. Expanding stakeholder scope to include workers and related professions would provide a fuller perspective. Comparative studies across developing countries using standardized measures could enhance generalizability. While this study provides valuable insights into stakeholder-related challenges in Türkiye’s construction sector, several methodological considerations should be noted. First, the analysis is based on Likert-type items, which are ordinal; however, treating such data as approximately continuous is widely accepted in SEM applications when distributions do not show substantial deviations, as was the case here. Second, EFA, CFA, and SEM were conducted on the same dataset. Given the sample size, dividing the data into separate subsamples would have reduced statistical power and potentially produced unstable estimates. To address this, we conducted comprehensive reliability and validity assessments and obtained a strong model fit, which strengthens confidence in the findings. Future research with larger datasets could compare alternative models or test hierarchical or bifactor structures to further confirm and refine the framework. Third, non-probability purposive sampling was used to ensure participation from professionals with relevant experience. Another methodological consideration is that the study did not include additional sensitivity analyses, such as alternative model specifications or subsample validations. Conducting such procedures typically requires larger samples or multiple theoretically grounded model variations, which were beyond the scope of the present design. Instead, model robustness was supported through strong global fit indices and consistent reliability and validity evidence. Future research using larger or multi-site samples could incorporate sensitivity tests—such as split-sample validation, bootstrapping-based comparisons, or alternative structural models—to further assess the stability and generalizability of the framework. Finally, the initial SMC list was intentionally inclusive to reflect the diversity of concepts in prior studies; subsequent filtering and factor-analytic procedures substantially reduced overlap and improved construct clarity. These considerations do not undermine the results but offer context for interpreting the findings and identifying opportunities for future research.