Governments in the Global South face ongoing challenges in digital modernization, ranging from weak technological infrastructure to entrenched bureaucratic cultures. This study draws on the Dynamic Capabilities View (DCV) and Contingency Theory to examine how digital transformation (DT) enhances flexibility in public service delivery when supported by organizational culture (OC) and technological infrastructure (TIF). Using the Cambodian public sector as a case study, survey data from 299 public officials were analyzed using Partial Least Squares Structural Equation Modelling (PLS-SEM) and Necessary Condition Analysis (NCA). The results show that DT improves flexibility in public services both directly and indirectly through OC and TIF, demonstrating that these factors are not merely contextual conditions but critical enablers of adaptive capacity. Importantly, NCA reveals that specific threshold levels of DT (≥2.59), OC (≥2.82–4.00), and TIF (≥3.37) constitute non-compensatory prerequisites for developing flexibility in public services. These findings extend the DCV by demonstrating that public-sector agility depends on the coordinated development of cultural and technological capabilities, and they refine Contingency Theory by highlighting that digital reform outcomes depend on the strategic alignment of institutional conditions. For policymakers, this study highlights the importance of simultaneous and coordinated investment in digital literacy, technological infrastructure, and cultural change to ensure that digital transformation initiatives lead to sustained improvements in public service responsiveness.
Digital transformation (DT) has become a cornerstone of contemporary governance, fundamentally reshaping how public services are designed, delivered, and evaluated through the strategic integration of digital technologies (Latupeirissa et al., 2024; Vial, 2021). By streamlining administrative processes and enhancing efficiency, DT aims to improve service accessibility, responsiveness, and citizen-centricity. However, the benefits of digital governance are unevenly distributed, particularly in developing contexts where structural, technological, and social constraints persist. A key concern is the widening digital divide, which disproportionately affects vulnerable and rural populations lacking access to reliable infrastructure and digital skills (Ahn & Chen, 2022; Bertot et al., 2016; Van Dijk, 2020). Without inclusive, context-sensitive implementation, digital transformation initiatives risk reinforcing rather than reducing existing inequalities in public service access.
These challenges are especially pronounced in the Global South. This study focuses on Cambodia as an illustrative case of a developing country characterized by a post-conflict legacy, limited administrative capacity, and entrenched bureaucratic norms. Such conditions make Cambodia a particularly relevant context for examining both the enablers and constraints of public-sector digital modernization (Ly, 2024a; Ly & Ly, 2023; Savuth & Sothea, 2023). Despite more than two decades of economic reform, Cambodia’s public sector continues to face low levels of technological readiness and underdeveloped information and communication technology (ICT) infrastructure, particularly in rural areas (Savuth & Sothea, 2023). Moreover, many civil servants lack formal training and confidence in using digital systems, further constraining effective implementation (Ly & Ly, 2023; Reddy et al., 2023). Although initiatives such as the 2019 E-Government Master Plan and the CamDX interoperability platform signal strong political commitment, bureaucratic rigidity and risk-averse organizational cultures continue to impede progress (Ly, 2025; Peeters & Campos, 2023).
Against this backdrop, this study focuses on flexibility in public services (FPS), defined as the capacity of public organizations to rapidly adapt service delivery models, administrative processes, and policy responses to changing citizen needs, environmental shifts, and unexpected crises (Rainey, 2009; van der Voet, 2014). The urgency of enhancing FPS became particularly evident during the COVID-19 pandemic, which exposed the limitations of paper-based systems and forced governments to innovate service delivery mechanisms under severe time constraints (AlNuaimi et al., 2022; Vigoda-Gadot & Mizrahi, 2024). While prior research acknowledges the importance of FPS for adaptive governance (Cedergren & Hassel, 2024), empirical evidence remains limited regarding how flexibility is systematically generated in resource-constrained public sectors, especially in Cambodia. Existing literature suggests that the success of digital transformation depends not only on the adoption of digital technologies but also on the organizational and infrastructural capacities that enable adaptation (Mettler et al., 2024; Rodríguez-González et al., 2023). In hierarchical and tradition-bound systems such as Cambodia’s public administration, cultivating a culture of innovation and adaptability is both necessary and challenging. While much of the digital transformation literature focuses on developed economies, relatively little attention has been paid to how digital transformation interacts with organizational culture (OC) and technological infrastructure (TIF) to jointly enable flexibility in developing-country contexts. This study addresses this gap by arguing that DT acts as a strategic driver that shapes both cultural and infrastructural conditions through top-down reform initiatives (Vial, 2021).
Accordingly, this study conceptualizes organizational culture and technological infrastructure as distinct yet interrelated mediating mechanisms through which digital transformation influences flexibility in public services. Organizational culture captures shared norms, learning orientation, and openness to change, while technological infrastructure reflects digital scalability, interoperability, and system integration. Together, these dimensions form complementary dynamic capabilities that allow public organizations to reconfigure resources in response to uncertainty. The study addresses the following research question: How does digital transformation influence flexibility in public services through organizational culture and technological infrastructure in Cambodia’s public sector? To answer this question, the study applies the Dynamic Capabilities View and Contingency Theory as analytical lenses rather than as objects of direct theory testing. It contributes in three ways. First, it clarifies the mediating roles of OC and TIF in translating digital initiatives into adaptive capacity. Second, it identifies both sufficient and necessary conditions, including threshold levels of culture and infrastructure, for achieving flexibility. Third, from a methodological perspective, it combines Partial Least Squares Structural Equation Modeling (PLS-SEM) with Necessary Condition Analysis (NCA), offering a nuanced understanding of both enabling pathways and binding constraints. These contributions provide theoretically grounded and policy-relevant insights for improving public-sector agility in resource-limited environments.
Literature reviewDigital transformation in public administration refers to the integration of digital technologies into governmental operations, fundamentally reshaping how public organizations function, interact with citizens, and deliver services (Mergel et al., 2019). This transformation is not limited to digitizing existing processes; rather, it involves a strategic shift toward leveraging digital tools to enhance efficiency, transparency, and adaptability in governance systems (Henriette et al., 2015). The overarching objective is to develop responsive and citizen-centered public services aligned with the demands of the digital age. Recent research highlights the role of DT in addressing complex societal challenges, including public health crises and economic disruptions, with digital innovations strengthening decision-making capacity and citizen engagement, as observed during the COVID-19 pandemic (Hess et al., 2020; Stoumpos et al., 2023). Within this context, flexibility in public services (FPS) represents a critical capability, reflecting the government’s ability to reallocate resources and adjust service delivery mechanisms in response to evolving citizen expectations. E-governance initiatives further enhance service accessibility, transparency, and administrative efficiency (Ly, 2024a; Ly & Ly, 2023). Digital transformation also promotes transparency and accountability by digitizing administrative processes and increasing access to government data, thereby enabling public scrutiny and reducing opportunities for corruption (Nambisan et al., 2019). Moreover, DT enhances policy agility by allowing governments to respond rapidly to environmental and societal changes (Ly, 2024a; Vial, 2021). Despite these potential benefits, the mechanisms through which DT translates into sustained flexibility outcomes remain insufficiently theorized. Existing studies often treat technology adoption as inherently transformative while overlooking the mediating roles of organizational culture and technological infrastructure. In developing contexts such as Cambodia, where bureaucratic rigidity and limited technological readiness persist, DT offers a particularly useful lens for examining how digital initiatives interact with organizational and infrastructural enablers to foster flexibility (Ly, 2025).
International examples such as Estonia’s e-government system (Kattel & Mergel, 2019) and South Korea’s advanced digital infrastructure (Chung et al., 2022; Lee-Geiller & Lee, 2022) demonstrate that public-sector flexibility is not merely the outcome of technological investment, but rather the result of aligning digital tools with adaptive governance norms and institutional capacity. By contrast, Cambodia continues to face challenges related to bureaucratic inertia, resource scarcity, and low levels of digital literacy, which constrain the translation of DT initiatives into improved flexibility in public services (Ly, 2024a, 2024b). Initiatives such as the CamDX interoperability platform aim to enhance inter-agency coordination and data exchange (Ly, 2025). However, their effectiveness depends on an organizational culture that fosters openness and collaboration, and a technological infrastructure that enables scalability and system integration (Hesmondhalgh et al., 2023; Ly, 2025; Sarwar et al., 2024). Accordingly, this study conceptualizes organizational culture (OC) and technological infrastructure (TIF) not as background conditions, but as active mediating mechanisms through which digital transformation influences flexibility in public services.
Technological infrastructure constitutes the backbone of digital transformation, enabling efficient public service delivery through reliable ICT systems, cloud computing, and cybersecurity mechanisms (Alvarenga et al., 2020). Without adequate TIF, the potential benefits of DT for public service flexibility cannot be fully realized. While e-governance systems can enhance accessibility and transparency, their effectiveness relies heavily on stable connectivity and robust data security frameworks (AlNuaimi et al., 2022; Sivarajah et al., 2017). Recent technological advancements, including AI-driven analytics and Internet of Things (IoT) applications, have further improved public-sector responsiveness, underscoring that TIF functions as a dynamic capability that enables scalability and resource reconfiguration (Maragno et al., 2023). Although cloud-based solutions offer significant scalability advantages, they also require strong data protection and regulatory compliance mechanisms (Arvind. et al., 2023; Qi et al., 2023). Similarly, AI- and IoT-based applications, such as chatbots and traffic management systems, automate administrative tasks and support urban planning but depend on real-time processing and system interoperability (Chen et al., 2023; Gil-Garcia et al., 2021). In Cambodia, persistent infrastructural challenges, including outdated systems and limited rural connectivity, exacerbate the digital divide and restrict equitable access to public services (Ly, 2024a, 2024b). These limitations constrain the public sector’s capacity to dynamically reconfigure resources, a core principle of the Dynamic Capabilities framework. More broadly, public organizations in Southeast Asia face resource constraints and skill shortages that hinder digital innovation (Henson, 2023; Ly, 2024a; Mergel, 2016). Public–private partnerships (PPPs) offer a viable mechanism for addressing these constraints by leveraging private-sector expertise to accelerate infrastructure development and reduce implementation costs (Alim & Polak, 2016; Yekimov et al., 2023). In the Cambodian context, such collaborations could play a critical role in expanding rural ICT infrastructure and supporting long-term capacity-building efforts.
Organizational culture represents the behavioral dimension of dynamic capabilities, shaping how individuals and institutions interpret change, share knowledge, and experiment with new technologies (Nambisan et al., 2019). Cultures that promote learning, cross-departmental collaboration, and openness to innovation enable employees to leverage digital tools for creative problem-solving and adaptive service delivery (Agger & Sørensen, 2018). In contrast, hierarchical and risk-averse cultures, common in many developing public administrations, often suppress innovation and hinder the successful implementation of digital transformation initiatives (Ly & Ly, 2023). In Cambodia, organizational culture plays a critical mediating role in the DT–FPS relationship by influencing how digital tools are accepted, adapted, and scaled across public institutions. Cultural adaptability, defined as the willingness to embrace change and support learning, complements technological scalability, which reflects the capacity to expand and integrate digital systems effectively. Together, these dimensions shape the public sector’s ability to convert digital investments into meaningful flexibility outcomes.
Flexibility in public administration is essential for responding to changing conditions and evolving citizen needs, particularly in developing countries such as Cambodia. This adaptability is closely aligned with Contingency Theory, which posits that flexible organizational structures can mitigate bureaucratic rigidity and improve governance effectiveness by enhancing policy formulation, decision-making, and service delivery (Cedergren & Hassel, 2024; Lenz & Eckhard, 2025). The COVID-19 pandemic underscored the importance of flexibility, as public administrations worldwide were compelled to rapidly adapt to unprecedented challenges, frequently relying on digital platforms to maintain operational continuity (Cedergren & Hassel, 2024; Dudau et al., 2023). In Cambodia, flexibility enables public institutions to overcome hierarchical constraints and respond to socioeconomic change, contributing to broader debates on governance in Southeast Asia. Integrating digital transformation with organizational culture and technological infrastructure reflects a contingency logic: the effectiveness of digital initiatives depends on their alignment with organizational and environmental conditions. From a dynamic capability’s perspective, this alignment facilitates continuous sensing, learning, and resource reconfiguration, which are necessary for adaptive public service delivery. Nevertheless, digital transformation in the public sector is often constrained by tensions between rigid administrative structures and the need for experimentation and flexibility (Selten & Klievink, 2024). This tension highlights the paradox of public-sector digitalization, in which digital tools can simultaneously reinforce bureaucratic control and enhance agility. This theoretical issue is explicitly examined in this study. To foster flexibility, Cambodia may benefit from targeted training programs and incentive mechanisms that reduce bureaucratic resistance and promote adaptability, aligned with dynamic performance management approaches (Bianchi & Hall, 2023; Looks et al., 2024). Moreover, inclusive governance models that empower communities and promote social justice can further enhance flexibility by incorporating diverse stakeholders into decision-making processes, thereby improving the representativeness and effectiveness of public policies (Sánchez-Soriano et al., 2024). Collectively, these insights reinforce the view that flexibility in public services is not merely an operational outcome, but a higher-order capability emerging from the interaction of digital, cultural, and infrastructural enablers.
Theoretical foundationThis study develops a theory-informed conceptual framework to explain how digital transformation (DT) affects flexibility in public services (FPS) through the mediating roles of technological infrastructure (TIF) and organizational culture (OC) in Cambodia’s public sector. The framework is grounded in two complementary theoretical perspectives: the Dynamic Capabilities View (DCV) and Contingency Theory, which together elucidate both the mechanisms and contextual conditions through which DT generates adaptive outcomes in public organizations. The Dynamic Capabilities View (DCV) (Teece et al., 1997) emphasizes an organization’s ability to sense opportunities and threats, seize them through timely and coordinated action, and reconfigure internal competencies to sustain performance under conditions of uncertainty and change (Mettler et al., 2024). Within this perspective, DT functions as a catalyst for capability renewal; however, its effectiveness depends on whether public organizations can reconfigure cultural and technological resources to respond to evolving demands. Contingency Theory, by contrast, posits that organizational effectiveness arises from the alignment between internal structures and external environmental conditions, rather than from any single optimal configuration (Donaldson, 2001). Together, these perspectives highlight that flexibility in public services does not emerge automatically from digital adoption, but rather from the alignment between digital capacity and institutional adaptability.
In the proposed model (see Fig. 1), digital transformation is conceptualized as a strategic driver that initiates both technological and cultural adaptation processes, rather than as a purely technical input. Governments may adopt digital tools to modernize service delivery; however, such tools acquire transformative capacity only when embedded within supportive institutional contexts that facilitate learning, collaboration, and scalability (Rodríguez-González et al., 2023; Vial, 2021). Accordingly, the model posits that organizational culture and technological infrastructure function as interconnected mediating mechanisms, shaping the behavioral and structural pathways through which DT influences flexibility in public services. Technological infrastructure is defined here as the structural capability that enables digital innovation and scalability. It encompasses information systems, interoperability mechanisms, and data management capacities that allow public organizations to deploy, expand, and secure digital operations effectively (Alvarenga et al., 2020). Rather than serving merely as a background resource, TIF operates as a dynamic capability that determines how efficiently digital initiatives are scaled and integrated into service delivery processes. In the Cambodian context, TIF influences the impact of DT by establishing minimum technological thresholds necessary for agility and system reconfiguration (Ly, 2024a, 2025), consistent with the logic of necessary conditions (Huang, 2019; Maragno et al., 2023). Conversely, weak or fragmented infrastructure can constrain DT outcomes by creating operational bottlenecks and data silos (Ly & Ly, 2023; Reddy et al., 2023).
Organizational culture represents the behavioral capability through which public institutions internalize and operationalize digital change. It shapes how employees interpret technological initiatives, accept innovation, and apply new tools to create public value (Al-Ruithe et al., 2019; Singh et al., 2021). In Cambodia’s bureaucratic environment, rigid hierarchies and risk-averse norms often discourage experimentation, thereby limiting the effective use of digital technologies (Ly, 2024a, 2025; Ly & Ly, 2023). In contrast, cultures characterized by openness, shared learning, and collaboration enhance the organization’s ability to translate digital inputs into flexible service outcomes. By positioning OC as a mediating mechanism, the framework underscores that technological readiness alone is insufficient; it must be complemented by cultural adaptability to develop dynamic capabilities. Flexibility in public services is therefore conceptualized as the outcome of this interaction, representing a higher-order dynamic capability. FPS reflects the public sector’s capacity to adjust policies, administrative processes, and resource configurations in response to changing citizen needs, crises, and policy priorities (Ongaro & Tantardini, 2023). Unlike earlier studies that treat flexibility as a byproduct of digital reform, this framework conceptualizes FPS as a deliberate strategic capability that emerges from the continuous alignment of digital, cultural, and structural elements within public organizations.
Empirically, the proposed framework is examined using Partial Least Squares Structural Equation Modeling (PLS-SEM) to assess the sufficiency of hypothesized relationships, complemented by Necessary Condition Analysis (NCA) to identify non-compensatory constraints. This methodological pairing reflects a theoretical rationale rather than a procedural choice. While PLS-SEM captures the extent to which DT, OC, and TIF collectively enable flexibility, NCA identifies the minimum conditions required for flexibility to emerge (Hair et al., 2022; Sarstedt et al., 2022). This dual-logic analytical design mirrors the integration of DCV and Contingency Theory by recognizing that capability development is simultaneously path-dependent and context-contingent. As such, the framework not only explains how digital transformation contributes to public-sector flexibility but also clarifies the institutional conditions under which such contributions become viable.
Hypotheses developmentDigital transformation is increasingly recognized as a critical factor in enhancing public service delivery through the adoption of digital technologies (Latupeirissa et al., 2024). This integration enables public institutions to improve efficiency and streamline administrative procedures. From a perspective of dynamic capabilities, digital transformation encompasses more than simple technology adoption; it signifies a strategic competency to reorganize organizational resources and processes to enhance responsiveness and adaptability (Vial, 2019). However, growing concerns regarding the digital divide indicate that, without inclusive planning, digital initiatives may marginalize vulnerable groups (Bertot et al., 2016; Latupeirissa et al., 2024; Van Dijk, 2020). The COVID-19 pandemic further underscored the urgent need for digital transformation to enhance resilience and flexibility in public services (AlNuaimi et al., 2022). Therefore, digital transformation enables public institutions to respond dynamically to change, which constitutes a core dimension of flexibility in public services (FPS). Based on this reasoning, this study proposes that:
H1 Digital transformation significantly affects flexibility in public services.
Chwiłkowska-Kubala et al. (2023) argue that digital transformation is not merely about implementing new technologies but also involves profound shifts in business models, strategies, and organizational culture. Dynamic capabilities theory suggests that these shifts are essential for enabling learning and adaptability, which are core cultural attributes of flexible organizations (Deep, 2023). Organizational culture can either facilitate or hinder digital transformation, thereby shaping its outcomes (Shoaib, 2022; Singh et al., 2021). When digital transformation aligns with a supportive culture, it reinforces collaboration, innovation, and organizational performance, whereas cultural resistance constrains progress (Alshammari et al., 2024; Cardoso et al., 2024; El Rashied, 2022). Similarly, in the public sector, risk-averse cultures may inhibit experimentation and flexibility (Ly, 2024a, 2025; Ly & Ly, 2023). In developing contexts such as Cambodia, where hierarchical norms dominate, digital transformation must first reshape cultural assumptions before translating into improved service flexibility. Thus, this study proposes that:
H2 Digital transformation significantly affects organizational culture in Cambodia.
Organizational culture shapes flexibility through communication, collaboration, and leadership practices (Syahruddin et al., 2020). Cultures that emphasize innovation, autonomy, and learning foster flexibility by enabling organizations to adapt to uncertainty and evolving citizen needs (Anning-Dorson, 2021; Kafetzopoulos & Katou, 2024). Adaptive cultures encourage strategic experimentation and empower employees to respond creatively to challenges (Ling et al., 2021; Olafsen et al., 2021; Ray & Pana-Cryan, 2021). In the public sector, this translates into greater agility and responsiveness in service design and delivery, as cultural adaptability constitutes a behavioral capability that enables FPS by encouraging initiative and cross-functional collaboration among public employees. Based on these arguments, this study suggests that:
H3 Organizational culture significantly affects flexibility in public services.
Digital transformation drives upgrades in technological infrastructure, including system architecture, cybersecurity, and data management tools (Alvarenga et al., 2020). From the perspective of dynamic capabilities, technological infrastructure represents a structural capability that enables organizations to scale, integrate, and secure digital operations, thereby transforming digital adoption into adaptive performance (Li et al., 2022; Verhoef et al., 2021). In the public sector, this has resulted in improved e-governance systems and resilience-oriented technologies that support continuous service delivery (Daugulis, 2023; Lin & Xie, 2023). As digital transformation expands, the demand for scalable, adaptable, and secure infrastructure continues to grow (Skare et al., 2023; Sudrajat, 2021). Hence, this study suggests that:
H4 Digital transformation significantly influences technological infrastructure.
Technological infrastructure enhances flexibility by improving data integration, communication, and responsiveness within and across public agencies (Hou, 2020). Flexible IT systems enable timely decision-making and adaptable process reconfiguration, which are essential for public service agility (Bazaras et al., 2023; Chester & Allenby, 2021). Robust digital infrastructure enables governments to maintain service continuity and responsiveness under volatile conditions (Elshaiekh et al., 2023; Ha, 2022). In public settings, flexible infrastructure acts as a foundational enabler of FPS by ensuring scalability and interoperability to accommodate evolving policy and citizen needs (Ly, 2025; Tong et al., 2022). Accordingly, this study proposes the following hypothesis:
H5 Technological infrastructure significantly influences flexibility in public services.
In hierarchical public organizations, cultural factors critically mediate the success of digital transformation. Dynamic capabilities theory suggests that adaptive cultures support the integration and reconfiguration of new digital tools, thereby yielding concrete flexibility outcomes (Ly, 2025; Rodríguez-González et al., 2023). Conversely, rigid and risk-averse cultures often suppress innovation and digital adoption (Chouki et al., 2018; Christensen et al., 2016). Cultures characterized by high power distance and low tolerance for uncertainty discourage employees from experimenting or sharing ideas, thereby weakening the adaptive impact of digital transformation (Rodríguez-González et al., 2023). In contrast, empowering leadership and supportive team environments foster engagement, experimentation, and digital confidence (Alshammari et al., 2024; Cortellazzo et al., 2019; Tushman & O'Reilly, 1996). Based on these notions, this study suggests that:
H6 Organizational culture mediates the relationship between digital transformation and flexibility in public services.
Digital transformation also requires adequate technological infrastructure to effectively operationalize change. Without sufficient infrastructure, the potential benefits of DT cannot be realized, as digital systems require minimum levels of interoperability, data quality, and cybersecurity (Lu et al., 2023; Nguyen et al., 2023). From a contingency perspective, technological infrastructure functions as a necessary condition that enables organizations to reconfigure and sustain new digital practices (Huang & Ren, 2024; Wang et al., 2024). Moreover, prior studies confirm that TIF mediates the relationship between DT and flexibility by facilitating data exchange, communication, and coordination across service networks (Bui & Le, 2023; Gong et al., 2020). Accordingly, the ability of digital transformation to enhance FPS depends on the presence of sufficient technological infrastructure that supports scalability and adaptability. Therefore, this study proposes the following hypothesis:
H7 Technological infrastructure mediates the relationship between digital transformation and flexibility in public services.
This study employed a quantitative, cross-sectional research design using a structured questionnaire to collect data on public-sector officials’ perceptions of digital transformation (DT), technological infrastructure (TIF), organizational culture (OC), and flexibility in public services (FPS). The empirical setting was Cambodia’s public sector, which is characterized by hierarchical administrative structures and ongoing digital governance reforms. This context enabled the study to capture the organizational and technological dynamics shaping flexibility outcomes under resource-constrained conditions. The target population comprised public-sector officials from both central and local government agencies actively involved in Cambodia’s digital transformation initiatives. Respondents included mid-level managers, technical officers, and administrative staff from various ministries, thereby providing diverse organizational perspectives and reducing the risk of leadership-level bias.
A five-point Likert scale (“1 = Strongly Disagree” to “5 = Strongly Agree”) was employed to capture perceptions of four latent constructs. The questionnaire consisted of separate sections dedicated to each construct. First, DT assessed the extent of digital platform adoption and its perceived impact. Second, TIF evaluated the availability, quality, and interoperability of ICT infrastructure. Third, OC examined organizational adaptability, learning capacity, and openness to change. Finally, FPS assessed the capacity to adapt processes, integrate technology, and deliver services flexibly. To ensure content validity and reliability, the instrument was pilot tested with a small group of Cambodian public officials (n = 20). Feedback helped clarify ambiguous wording, confirmed cultural relevance, and improved item clarity prior to final deployment. The finalized questionnaire was distributed online via professional networking channels and social media platforms (Facebook and Telegram), which are widely used among Cambodian civil servants for inter-agency communication.
Given this mode of dissemination, the study adopted a quota-based, stratified convenience sampling approach, rather than a purely random design. The sample was stratified by administrative level (central versus local government) and ministry type (service-oriented versus administrative) to ensure coverage of key subgroups and reduce sampling bias (Taherdoost, 2016). Although participation was voluntary and self-selected, stratification was applied during recruitment to enhance group comparability and approximate representativeness. Recognizing the potential for coverage and self-selection biases associated with social media recruitment, the survey targeted official government groups verified by ministry communication units, and follow-up reminders were used to ensure balanced representation from both urban and provincial respondents.
The sample size was established based on Soper (2023) a priori power analysis, which indicated that 207 participants were sufficient for a model consisting of 20 observable variables and four latent constructs, assuming a statistical power of (1 − β) = 0.95 at a significance level of α = 0.05, with an anticipated effect size of λ = 0.3. To offset potential attrition or incomplete responses, the distribution target was increased by approximately 50 %. Data were screened for missing values, outliers, and normality using SPSS version 27 prior to structural modeling. Ethical standards were upheld by ensuring anonymity and voluntary participation, in accordance with governance research ethics (Ongaro & Tantardini, 2023). Of the 310 questionnaires distributed, 299 valid responses were obtained, yielding a response rate of 96.4 %, which exceeded the minimum requirement and supported statistical robustness (see Table 1).
The study employed Partial Least Squares Structural Equation Modeling (PLS-SEM) in combination with Necessary Condition Analysis (NCA) to examine both sufficient and necessary relationships among the constructs. This analytical strategy is consistent with the theoretical framework, in which the Dynamic Capabilities View emphasizes sufficiency logic, while Contingency Theory highlights necessity logic. PLS-SEM is particularly appropriate for analyzing complex causal models with mediating relationships and moderate sample sizes (Hair et al., 2022). In contrast, NCA enables the identification of non-compensatory constraints, such as minimum levels of organizational culture or technological infrastructure required for flexibility to emerge (Dul, 2016). The integration of PLS-SEM and NCA enhances methodological rigor by capturing both enabling pathways and binding constraints, thereby aligning the empirical analysis with the study’s focus on dynamic and context-contingent capabilities. While acknowledging that external factors, such as political dynamics, may influence perceptions of digital transformation, the analysis deliberately focuses on internal organizational variables that directly align with the study’s objectives. As such, the methodological framework strengthens the validity and contextual relevance of the findings for Cambodia and provides a replicable analytical approach for examining public-sector digital transformation in other developing-country contexts.
MeasuresThe questionnaire consisted of 20 measurement items addressing the study’s core constructs (see Table 2). All measurement items were adapted from previously validated scales identified through a comprehensive literature review and contextually refined to fit the Cambodian public-sector setting. To avoid redundancy and ensure conceptual consistency, the items were carefully reviewed and adjusted to reflect the institutional, cultural, and administrative characteristics of Cambodia’s public administration. A confirmatory factor analysis (CFA) was conducted to validate the measurement instrument. CFA is a theory-driven statistical technique used to identify the latent factors underlying observed variables and to assess the extent to which the measurement model fits the data (Byrne, 2013). Specifically, CFA evaluates the strength and significance of the relationships between observed indicators and their corresponding latent constructs, thereby establishing construct validity. During data cleaning, the study systematically screened for inconsistent or patterned response behaviors and found no evidence of such issues. The high internal consistency reliability of the constructs, as indicated by Cronbach’s alpha values exceeding 0.85 for all constructs, further suggests that response bias attributable to item wording or measurement design was not a significant concern.
Measurement items.
| 1. Digital Transformation (DT) |
| 1-Our organization has implemented digital platforms for public service delivery (Vial, 2019) |
| 2-Digital tools and technologies have enhanced our operational efficiency (Verhoef et al., 2021) |
| 3-We use digital technologies to streamline processes and improve decision-making (Chanias et al., 2019) |
| 4-Digital technologies align with our goals to improve citizen-centered services (Ly, 2024a) |
| 5-Our organization's digital transformation is often hindered by employees' digital skill gaps (Rodríguez-González et al., 2023) |
| 2. Technological Infrastructure (TIF) |
| 1-IT services are reliable and consistently available (Xin et al., 2015) |
| 2-Maintenance of hardware and software systems is carried out effectively (Kadnichanskiy et al., 2020) |
| 3-Cybersecurity measures and data protection policies are adequate (Norris & Mateczun, 2023) |
| 4- ICT infrastructure supports seamless integration of e-services (Ly, 2024a; Ly & Ly, 2023) |
| 5-Insufficient rural ICT connectivity limits our digital service delivery (Rodríguez-González et al., 2023) |
| 3. Organizational Culture (OC) |
| 1-Our organization promotes a culture of innovation and adaptability (Denison, 1996) |
| 2-Employees are encouraged to experiment with new digital tools and technologies (Schein, 2010) |
| 3-The leadership supports initiatives to enhance employee digital literacy (Quinn, 2011) |
| 4-Bureaucratic resistance hinders openness to digital innovation in our organization (Rodríguez-González et al., 2023) |
| 5-Our culture fosters collaboration to overcome hierarchical barriers to change (Ly, 2024a; Ly & Ly, 2023) |
| 4. Flexibility in Public Services (FPS) |
| 1-Our organization can quickly adapt to changes in public demands and service needs (Rainey, 2009) |
| 2-Our policies are flexible enough to accommodate technological changes and innovations (van der Voet, 2014) |
| 3-The organizational structure allows for quick decision-making and service adjustments (Tushman & O'Reilly, 1996) |
| 4-Our processes adapt to new digital platforms (Ly, 2024a; Ly & Ly, 2023) |
| 5-Centralized decision-making limits our ability to respond to local service needs (Rodríguez-González et al., 2023) |
This study employed a dual-logic analytical approach combining Partial Least Squares Structural Equation Modeling (PLS-SEM) and Necessary Condition Analysis (NCA) to comprehensively examine both sufficient and necessary causal relationships among the four constructs. PLS-SEM was implemented using SmartPLS 4 to test the structural model and assess the hypothesized relationships among digital transformation (DT), organizational culture (OC), technological infrastructure (TIF), and flexibility in public services (FPS). This technique was selected for its ability to analyze complex causal models with multiple mediating relationships and moderate sample sizes, particularly in emerging and resource-constrained research contexts (Hair et al., 2022; Sarstedt et al., 2020).
Consistent with the two-step procedure proposed by Anderson and Gerbing (1988), the analysis first evaluated the measurement model, focusing on indicator reliability, convergent validity, and discriminant validity. Subsequently, the structural model was assessed to test the hypothesized relationships and examine mediating effects (Hair et al., 2022; Henseler et al., 2016). The PLS-SEM estimation was based on 299 valid observations and employed 10,000 bootstrap resamples to generate robust, non-parametric path estimates and confidence intervals, thereby enhancing the statistical robustness of the findings (Ramayah et al., 2018).
Necessary Condition Analysis (NCA) was incorporated as a complementary analytical technique to PLS-SEM to identify relationships that are necessary but not sufficient for the outcomes of interest (Dul, 2016). While PLS-SEM explains how predictors collectively contribute to outcomes through sufficiency logic, NCA determines the minimum threshold levels below which desired outcomes, such as flexibility in public services (FPS), cannot occur, reflecting necessity logic. This distinction is particularly relevant in public-sector transformation contexts, where certain baseline conditions (e.g., minimum levels of technological infrastructure or cultural adaptability) must be present before digital reforms can generate flexibility (Richter et al., 2020). Thus, NCA enables the identification of non-compensatory bottlenecks, highlighting the must-have conditions required for FPS. The NCA procedure followed Dul et al. (2020) and was implemented using both the ceiling envelopment–free disposal hull (CE-FDH) and ceiling regression–free disposal hull (CR-FDH) techniques. Effect sizes (d) exceeding 0.10 were interpreted as indicating meaningful necessary effects, and accuracy indicators (R² and p-values) were calculated using permutation tests.
Accordingly, the combined application of PLS-SEM and NCA offers both explanatory and diagnostic insights: PLS-SEM identifies which factors influence flexibility and the strength of their effects, whereas NCA determines the minimum levels of enabling conditions required for flexibility to emerge. This dual-logic framework aligns closely with the Dynamic Capabilities View (DCV), which emphasizes capability sufficiency, and Contingency Theory, which highlights contextual necessity. Consequently, the integrated approach enhances both theoretical precision and practical relevance, enabling policymakers to distinguish between “nice-to-have” and “must-have” determinants of public-sector flexibility.
Measurement modelAll constructs in the model were measured reflectively and evaluated in accordance with established criteria for reflective measurement models, including internal consistency reliability, convergent validity, and discriminant validity. Reliability was assessed using Cronbach’s alpha and composite reliability (CR) for all survey items. As reported in Table 3, all reliability coefficients exceeded the recommended threshold of 0.70 (Hair et al., 2022). The CR and average variance extracted (AVE) values further confirmed the internal consistency and reliability of the measurement scales. Convergent validity was supported, as all standardized factor loadings exceeded 0.70 and all AVEs exceeded the recommended minimum of 0.50 (Fornell & Larcker, 1981), indicating that the constructs explained a substantial proportion of the variance in their indicators. Discriminant validity was assessed using the heterotrait–monotrait (HTMT) ratio of correlations (Henseler et al., 2015; Ringle et al., 2023). As shown in Table 4, all HTMT values were well below the conservative threshold of 0.90, and the bootstrapped confidence intervals from 10,000 resamples did not exceed 1.00. These results indicate that discriminant validity was adequately established and does not pose a concern for the interpretation of the structural model (Franke & Sarstedt, 2019).
Factor loadings, reliability, and validity.
The structural model was analyzed after the validity and reliability of the constructs had been established. Prior to estimating the structural relationships, multicollinearity was assessed using variance inflation factors (VIFs) according to the procedure recommended by Hair et al. (2011). For PLS-SEM, acceptable VIF values typically range from 0.20 to 5.00; values outside this range indicate potential multicollinearity. In this study, the VIF values ranged from 1.60 to 3.10, indicating that multicollinearity was not a concern. Subsequently, common method bias (CMB) was assessed using Harman’s single-factor test. The results indicated that a single factor accounted for 42.23 % of the total variance, which is below the 50 % threshold suggested by Podsakoff et al. (2003). These findings suggest that common method bias did not pose a significant threat to the validity of the results.
The quality of the structural model was assessed by examining the predictive performance of the endogenous constructs using multiple evaluation indices, including the coefficient of determination (R²), cross-validated redundancy (Q²), standardized path coefficients (β), and path significance. Standardized path coefficients were used to evaluate the magnitude and direction of the hypothesized relationships (Gallardo-Vázquez & Sánchez-Hernández, 2014). The R² values were employed to assess the model’s explanatory power, with values of 0.10 or higher considered acceptable in behavioral and social science research (Falk & Miller, 1992).
Cross-validated redundancy (Q²) was used to assess the predictive relevance of the endogenous constructs, with Q² values greater than 0.00 indicating meaningful predictive capability (Castro & Roldán, 2013). As reported in Table 5, the R² value for organizational culture (OC) was 0.165, indicating that digital transformation (DT) explained 16.5 % of the variance in OC, while the corresponding Q² value was 0.102, suggesting moderate predictive relevance. For technological infrastructure (TIF), the R² value was 0.448, indicating that DT accounted for 44.8 % of the variance, and the Q² value was 0.287, reflecting strong predictive relevance. For flexibility in public services (FPS), the R² value reached 0.496, suggesting that the combined effects of DT, OC, and TIF explained 49.6 % of the variance, while the Q² value of 0.298 indicated strong predictive relevance.
The standardized root mean square residual (SRMR) was used to assess potential model misspecification in PLS-SEM by measuring the discrepancy between observed and model-implied correlations (Hu & Bentler, 1999; Kenny, 2020). An SRMR value below 0.10 is generally considered indicative of an acceptable model fit (Hu & Bentler, 1998; Kara et al., 2022; Worthington & Whittaker, 2006). In this study, the model demonstrated an acceptable overall fit, with an SRMR value of 0.09.
Subsequently, the structural model was assessed using bootstrapping with 10,000 subsamples to test the proposed hypotheses. All hypothesized relationships were supported, as summarized in Table 5. Specifically, the results for H1 indicate that digital transformation (DT) exerts a significant positive effect on flexibility in public services (FPS) (β = 0.239, t = 3.341, p = 0.001), suggesting that organizations engaged in digital transformation initiatives exhibit higher levels of service flexibility. Consistent with H2, DT was found to have a significant positive influence on organizational culture (OC) (β = 0.406, t = 5.779, p < 0.001), indicating that digital transformation efforts contribute to the development of more adaptive and supportive organizational cultures. In line with H3, organizational culture demonstrated a positive and significant effect on FPS (β = 0.337, t = 4.522, p < 0.001), implying that culturally supportive and dynamic environments enhance public service flexibility. Regarding technological infrastructure, the results for H4 reveal a strong positive relationship between DT and TIF (β = 0.669, t = 19.039, p < 0.001), indicating that digital transformation plays a critical role in strengthening and upgrading organizational technological infrastructure. Finally, H5 shows that TIF has a significant positive effect on FPS (β = 0.265, t = 3.748, p < 0.001), underscoring the importance of robust and scalable technological infrastructure for enabling flexible public service delivery.
Furthermore, the mediation analysis reported in Table 6 examines the direct, indirect, and total effects of digital transformation on public service flexibility. The total effect of DT on FPS was statistically significant (β = 0.554, t = 10.073), indicating a strong overall influence when both direct and mediated pathways are considered. The direct effect of DT on FPS remained significant (β = 0.239, t = 3.341), demonstrating that digital transformation independently contributes to public service flexibility. Additional analyses assessed the mediating roles of organizational culture and technological infrastructure. The indirect effect of DT on FPS through organizational culture (H6) was significant (β = 0.137, t = 4.101, p < 0.001), with a 95 % confidence interval of 0.072–0.200, confirming that OC partially mediates the relationship between DT and FPS. This finding suggests that digital transformation enhances organizational culture, thereby contributing to greater flexibility in public services. Similarly, the indirect effect of DT on FPS through technological infrastructure (H7) was significant (β = 0.177, t = 3.543, p < 0.001), with a confidence interval of 0.084–0.282. This result indicates that technological infrastructure also functions as a key mediating mechanism through which digital transformation improves service flexibility. Because both direct and indirect effects of DT on FPS were statistically significant, the results support partial mediation rather than full mediation (Baron & Kenny, 1986; Zhao et al., 2010). This implies that while organizational culture and technological infrastructure explain a substantial portion of the effect of digital transformation on flexibility, DT continues to exert a direct influence on FPS. Collectively, these findings underscore the multifaceted role of digital transformation, highlighting organizational culture and technological infrastructure as critical pathways through which digital initiatives enhance flexibility in public services.
Moreover, Necessary Condition Analysis (NCA) was conducted alongside PLS-SEM to further elucidate the relationships among digital transformation (DT), organizational culture (OC), technological infrastructure (TIF), and flexibility in public services (FPS). This analysis aims to identify the fundamental prerequisites that must be present for a specific outcome to occur (Dul et al., 2024; Manchiraju et al., 2024; Tiwari et al., 2024). The NCA effect size was calculated using the ceiling envelopment–free disposal hull (CE-FDH) technique to determine necessary conditions among the constructs (Abner et al., 2023; Dul et al., 2023). The results of the CE-FDH assessment are summarized in Table 7, and Fig. 2 presents the corresponding scatter plot.
According to Dul et al. (2024), an effect size below 0.10 combined with a p-value greater than 0.05 indicates that no necessary condition exists between the independent and dependent variables. For a construct to be classified as a necessary condition, three criteria must be satisfied: (i) a clear theoretical rationale, (ii) a positive effect size, and (iii) a statistically significant p-value (p < 0.05) (Dul et al., 2020). The NCA results reported in Table 7 indicate that DT, OC, and TIF all constitute necessary conditions for FPS, as each construct exhibited a p-value below 0.05.
Among the constructs, organizational culture demonstrated the largest effect size (CE-FDH = 0.286), followed by technological infrastructure (CE-FDH = 0.241) and digital transformation (CE-FDH = 0.075). These findings suggest that while all three constructs are necessary for achieving flexibility in public services, organizational culture and technological infrastructure play particularly influential roles. The relatively smaller effect size of DT indicates that although digital transformation is a necessary prerequisite, it is not sufficient on its own to achieve high levels of flexibility in public services. Rather, a supportive organizational culture and robust technological infrastructure are critical enabling conditions for translating digital transformation initiatives into meaningful outcomes in flexibility.
Moreover, the bottleneck analysis technique was applied to determine the threshold levels of digital transformation (DT), organizational culture (OC), and technological infrastructure (TIF) required to achieve specific levels of flexibility in public services (FPS).
As reported in Table 8, achieving 100 % FPS requires minimum threshold levels of DT ≥ 2.650, OC ≥ 4.000, and TIF ≥ 3.366. Different combinations of these factors are therefore necessary to achieve lower FPS levels, reflecting the non-compensatory nature of the relationships.
Bottleneck table.
For example, to achieve a 60 % FPS level, the minimum required thresholds are DT = 2.646, OC = 2.822, and TIF = 3.366. Notably, at 0 % and 10 % FPS, organizational culture was not identified as a necessary condition, whereas technological infrastructure became necessary at 10 % FPS and above. Digital transformation also emerged as a necessary condition starting at the 10 % FPS level, highlighting its role in supporting the initial enhancement of public service flexibility. The threshold values for DT remained relatively stable across the mid-range levels of FPS, suggesting that once a baseline level of digital transformation is achieved, further increases in FPS do not require substantial additional investment in DT. In contrast, organizational culture and technological infrastructure become increasingly critical as higher levels of FPS are targeted.
For instance, the OC threshold increased notably at the 60 % FPS level to 2.822 and continued to rise to 4.000 at 100 % FPS. Similarly, TIF exhibited a consistent threshold requirement, beginning at 60 % FPS and remaining at 3.366 at higher levels. Overall, this bottleneck analysis underscores the importance of progressively achieving minimum threshold levels of DT, OC, and TIF to enhance flexibility in public services. While digital transformation is essential at the early stages, a supportive organizational culture and robust technological infrastructure become increasingly vital for achieving and sustaining higher levels of public service flexibility.
DiscussionThis study investigated how digital transformation (DT) enhances the flexibility of public services (FPS) in Cambodia, drawing on the combined perspectives of Contingency Theory and the Dynamic Capabilities View (DCV). By utilizing PLS-SEM and Necessary Condition Analysis (NCA), the study identified both relational effects and critical threshold conditions, demonstrating that specific levels of technological infrastructure (TIF) and organizational culture (OC) are essential for achieving flexibility. This dual analytical approach enhances theoretical precision by distinguishing between factors that are sufficient to promote FPS and those that are necessary for its emergence and sustainability.
Consistent with the DCV, the findings support the hypothesis that DT directly improves FPS (H1) by enabling public organizations to sense, respond to, and adapt to environmental change through digitized procedures and data-driven decision-making. This result aligns with prior research showing that DT enhances public responsiveness during crises (AlNuaimi et al., 2022) and promotes operational agility (Skare et al., 2023). In Cambodia, platforms such as CamDX and mobile e-government applications exemplify this shift toward resource reconfiguration to deliver more citizen-centric services. However, the NCA results indicate that DT alone is insufficient; minimum threshold levels of both OC and TIF must be achieved for digitalization to translate into meaningful flexibility. From a contingency perspective, the results reaffirm that the effectiveness of DT depends on its alignment with contextual conditions. In Cambodia’s rural areas, limited broadband infrastructure and low levels of digital literacy significantly constrain scalability (Reddy et al., 2023), supporting Latupeirissa et al. (2024)’s warning that digital reforms may exacerbate exclusion if infrastructural disparities persist. This finding extends Contingency Theory by demonstrating that, in resource-constrained contexts, digital transformation functions as a conditional capability, generating flexibility only when institutional, technological, and social infrastructures evolve in a coordinated manner. The observed digital divide thus underscores a critical limitation of DT in developing countries: reforms that neglect foundational readiness may reinforce inequality rather than promote inclusive modernization.
The study further confirms that DT positively influences OC (H2), consistent with prior studies (e.g., Alshammari et al., 2024; Chwiłkowska-Kubala et al., 2023; Shoaib, 2022), which suggest that digitalization fosters transparency, decentralization, and innovation. In Cambodia, where bureaucratic traditions remain deeply embedded, DT appears to be gradually encouraging shifts toward collaboration and participatory norms, particularly within digitally progressive ministries. The positive relationship between OC and FPS (H3) is also supported by Kafetzopoulos and Katou (2024), who found that flexibility-oriented cultures enhance strategic agility and decision-making quality. Similarly, Ray and Pana-Cryan (2021) and Syahruddin et al. (2020) predicted that adaptive cultures strengthen service responsiveness—findings echoed in the present study, in which agencies with stronger cultural engagement demonstrated greater flexibility.
Similarly, the role of DT in strengthening TIF (H4) was confirmed, aligning with Alvarenga et al. (2020) and Daugulis (2023), who argued that digital reforms stimulate investment in technological architecture. Accordingly, Cambodian public organizations that have advanced digital reforms, such as cloud-based systems and integrated data platforms, exhibit greater operational agility. However, the NCA results reveal a critical insight: TIF exhibits a threshold effect, meaning that unless a minimum level of interoperability and data infrastructure is achieved, incremental increases in DT have a limited impact on FPS. This finding complements the contingency perspective by demonstrating that infrastructural sufficiency is a prerequisite rather than a performance enhancer. Moreover, the positive effect of TIF on FPS (H5) corroborates prior findings by Bazaras et al. (2023) and Chester and Allenby (2021)), who claim that flexible IT environments enhance communication, data sharing, and responsiveness. In Cambodia, agencies equipped with modernized infrastructure demonstrate higher adaptability, whereas under-resourced institutions struggle to respond effectively to emerging public needs.
The mediating roles of OC (H6) and TIF (H7) provide robust evidence that flexibility in public services emerges through the joint alignment of behavioral and structural capabilities. These mediation effects confirm that the effectiveness of DT depends on both soft infrastructure (organizational culture) and hard infrastructure (technology). This supports earlier findings (e.g., Bui & Le, 2023; Huang & Ren, 2024; Rodríguez-González et al., 2023) that culture and infrastructure are not passive enablers but active mediators that convert digital investments into adaptive capacity. From a DCV perspective, OC and TIF together function as orchestrating mechanisms that enable public organizations to reconfigure resources under conditions of uncertainty. From a contingency lens, they define the institutional fit required for digital transformation to yield flexibility rather than rigidity.
While many findings align with existing literature, the results also reveal important paradoxes. For example, Christensen et al. (2016) and Julagasigorn et al. (2025) cautioned that in hierarchical bureaucracies, DT may strengthen rather than decentralize administrative control, as digital tools become instruments of surveillance. Evidence from several Cambodian public organizations supports this paradox: digital systems are sometimes deployed to monitor compliance rather than encourage innovation. Similarly, Ly (2025) and Sokhea and Sophea (2019) highlighted how ICT skill deficits and cultural inertia among senior officials constrain the transformative potential of DT. This paradox of “digital rigidity” suggests that digitalization, when not accompanied by cultural and capability development, risks reproducing bureaucratic behaviors through new technological channels. This insight represents an important theoretical contribution, extending DCV interpretations into the domain of public governance.
Finally, the NCA findings deepen these insights by distinguishing necessary from sufficient enablers of flexibility. Specifically, the results indicate that minimum threshold levels of DT (≥ 2.59), organizational culture (≥ 2.82 for moderate flexibility, increasing to 3.19–4.00 for high flexibility), and technological infrastructure (≥ 3.37) are required for FPS to materialize. Below these thresholds, flexibility cannot emerge regardless of improvements in other factors, underscoring the non-compensatory nature of these conditions. Beyond these minimum levels, further gains in FPS diminish unless OC and TIF are developed simultaneously. This joint dependency demonstrates that culture and infrastructure operate in complementary rather than independent ways, thereby advancing theoretical understanding of dynamic capability complementarity. Conceptually, the findings suggest that public organizations cannot substitute technological investment for cultural adaptability; both are indispensable for achieving sustained agility and resilience in public service delivery.
ConclusionThis study demonstrates how digital transformation (DT) improves flexibility in public services (FPS) in Cambodia’s public sector by operating through the mediating roles of organizational culture (OC) and technological infrastructure (TIF). Grounded in the Dynamic Capabilities View (DCV) and Contingency Theory, the findings reveal that DT enhances flexibility not merely as a technological intervention, but as a strategic capability whose effectiveness depends on institutional context and internal alignment. Both OC and TIF function as behavioral and structural enablers that translate digital initiatives into adaptive public service delivery. Necessary Condition Analysis (NCA) indicates that minimum threshold levels of DT (≥ 2.59), OC (≥ 2.82–4.00), and TIF (≥ 3.37) are required for flexibility to emerge, highlighting that organizational culture and technological infrastructure must simultaneously reach sufficient levels. These findings emphasize the interdependence between digital and organizational capabilities.
From a practical perspective, policymakers should prioritize short-term interventions, such as digital literacy training, the development of interoperable systems, and the establishment of innovation “sandboxes,” while simultaneously pursuing long-term institutional reforms to foster an adaptive organizational culture and ensure sustained investment in digital infrastructure. Overall, the study offers an integrated, theory-informed framework that clarifies the necessary and sufficient conditions for digital transformation to generate genuine public-sector agility in developing-country contexts.
LimitationsThis study has certain limitations that open important avenues for future research. First, the cross-sectional research design restricts strong causal inference regarding the relationships between digital transformation (DT), the mediating variables, and flexibility in public services (FPS). A longitudinal research design would better capture how these relationships evolve over time and allow assessment of the long-term sustainability of digital transformation initiatives. Second, the study relies on self-reported data, which may introduce perceptual and common method biases. Although Harman’s single-factor test indicated no substantial threat, future studies could strengthen methodological rigor by triangulating survey responses with objective performance indicators or by applying procedural remedies, such as temporal separation, multiple data sources, or marker variables, to further mitigate such biases. Third, the study’s focus on Cambodia’s public sector limits the external generalizability of the findings. While Cambodia provides a valuable empirical context for examining digital governance in developing-country settings, comparative studies across other Global South countries would enhance external validity and reveal contextual variations in institutional capacity, administrative traditions, and reform trajectories. In addition, factors such as political dynamics, donor influence, and institutional autonomy, which are addressed only at a general level in this study, warrant more explicit examination in future research. Employing multilevel modeling approaches could help uncover their cross-level effects on digital transformation outcomes and public service flexibility. Finally, the exclusive use of quantitative survey methods may not fully capture the cultural, organizational, and institutional nuances underlying digital transformation processes. Future research could adopt mixed-method designs, integrating interviews, focus groups, or in-depth case studies to explore the microfoundations of dynamic capabilities and the role of digital ecosystems in shaping public-sector transformation. Cross-country and regional comparisons would further advance a more comprehensive theory of public-sector digital transformation in resource-constrained environments.
FundingThe authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
CRediT authorship contribution statementBora Ly: Writing – review & editing, Writing – original draft, Validation, Supervision, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Romny Ly: Writing – review & editing, Software, Formal analysis, Data curation, Conceptualization. Sokhom Ma: Writing – review & editing, Writing – original draft, Resources, Data curation.
