Building Digital Governance Capacity for Digital Transformation in Public Administration: Evidence from Lima, Peru

Abstract

Digital transformation has become a strategic pillar of modernization in public administration; however, evidence from emerging economies shows that technology implementation alone does not guarantee institutional transformation. This study examines the structural relationships among digital competencies, change management, technology adoption, and digital transformation in public administration institutions in Lima, Peru, incorporating Digital Inclusion Practices (DIP) as a moderating variable. A quantitative, non-experimental, cross-sectional design was applied to a sample of 358 public servants working in ministries, national agencies, regional administrative units, and municipal governments located in Lima. Data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results indicate that digital competencies and change management are positively associated with technology adoption, and that change management and technology adoption show significant positive relationships with digital transformation. Technology adoption partially mediates the relationships between digital competencies and digital transformation, and between change management and digital transformation. Additionally, digital competencies show a direct and statistically significant, although weak, relationship with digital transformation. The moderating relationship involving DIP was not statistically significant. These findings suggest that digital governance in emerging public administrations may depend less on individual skills alone and more on structured institutional processes that support effective technology adoption, strategic change management, and institutional modernization.

1. Introduction

Digital transformation has consolidated itself as one of the central pillars of state modernization in the 21st century. Governments face increasing pressure to improve administrative efficiency, strengthen transparency, and optimize the delivery of public services through the strategic use of digital technologies. In this context, digitalization not only implies incorporating technological tools, but also redefining processes, structures, and organizational dynamics to generate public value and respond to environments characterized by high uncertainty and changing citizen demands (Norling, 2025; Tinjan, 2025). Digital transformation thus becomes a strategic governance process that impacts both internal management and the relationship between the State and society. In this sense, digital transformation has become a central element of the contemporary public administration agenda, as it redefines how government institutions design, manage, and deliver public services.

However, recent evidence indicates that the mere incorporation of technologies does not guarantee effective processes of institutional transformation. Various studies show that the outcomes of digitalization largely depend on the human and organizational capabilities that sustain these processes (Aristovnik et al., 2025; Santos, 2024). In the public sector, factors such as civil servants’ digital competencies, organizational change management, and the willingness to adopt new technologies emerge as key conditions for digitalization to translate into structural and sustainable improvements (Adewumi & Abasilim, 2024; Barodi et al., 2024). In this way, digital transformation should not be understood as a purely technological phenomenon, but as a process that integrates individual, organizational, and strategic dimensions.

In Latin America, the challenges are even more complex due to structural inequalities and institutional limitations that characterize many administrative systems. Although digital government initiatives have been developed in the region, significant gaps persist in digital capabilities, institutional coordination, and change management (Espina-Romero, 2025). In urban contexts such as Lima, Peru, public administration modernization faces tensions derived from traditional bureaucratic structures and heterogeneous levels of digital readiness. Added to this is the need for stronger digital inclusion practices, understood as institutional actions aimed at reducing inequalities in access, use, and effective benefit from digital technologies. These practices may support the scope and inclusiveness of digital transformation policies (Singh & Chobotaru, 2022; Tran Pham & Le Hoang Thuy To Nguyen, 2024).

Peru is a relevant setting for this study because digital transformation in public administration is supported by a formal national governance framework. The Digital Government Law establishes the governance framework for digital identity, digital services, interoperability, digital security, and data management across public administration entities at the three levels of government. Likewise, Emergency Decree No. 006-2020 created the National Digital Transformation System, which organizes public administration activities and promotes coordination with the private sector, civil society, academia, and citizens (Presidencia del Consejo de Ministros, 2020). The National Digital Transformation Policy to 2030 further frames digital transformation as a continuous, strategic, and cultural change process aimed at generating public value through digital technologies and data (Presidencia del Consejo de Ministros, 2018).

Lima is especially relevant within this framework because it is Peru’s capital and main administrative center, concentrating ministries, national agencies, regional administrative units, and municipal governments. It also represents the country’s largest urban area, with more than 10.4 million inhabitants and approximately 30.4% of the national population (Instituto Nacional de Estadística e Informática, 2025). However, these institutions do not share the same mandates, resources, technological infrastructure, or implementation capacity. For this reason, Lima offers a useful empirical setting for analyzing how digital competencies, change management, technology adoption, and digital inclusion practices operate within public administration institutions exposed to the same national digital transformation agenda but marked by heterogeneous levels of institutional readiness.

At the theoretical level, the literature has partially addressed these dimensions through approaches such as the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT), which explain technology adoption based on individual perceptions (Aranyossy, 2022; Aristovnik et al., 2025). Likewise, dynamic capabilities theory has been adapted to the public sector to analyze how organizations develop competencies that allow them to respond to changing environments (Santos, 2024). However, gaps still exist in integrating these frameworks within structural models that articulate digital competencies, change management, and technology adoption as explanatory mechanisms of digital transformation in the Latin American public sector. In addition, the literature has paid less attention to the role of digital inclusion practices as institutional mechanisms that may condition the relationship between technology adoption and digital transformation.

Although previous studies have examined the relationships between digital skills, technology adoption, change management, and digital transformation, these relationships have often been analyzed as separate explanatory factors (Verhoef et al., 2021; Vial, 2019). This limits the theoretical understanding of how public institutions transform digital resources into broader modernization outcomes. The present study addresses this limitation by framing these relationships as part of a Digital Governance Capacity mechanism. In line with prior research, DCs provide the human capability required to effectively use and adapt digital technologies (van Laar et al., 2017), while CM provides the organizational support needed to reduce resistance and facilitate the implementation of new digital practices.

TA operates as the mechanism through which digital tools become embedded in administrative routines and organizational processes (Davis, 1989; Venkatesh et al., 2003), and DIP represents the inclusion-oriented dimension that seeks to ensure equitable participation in digital environments and the generation of public value. Therefore, the theoretical novelty of this study does not lie in proposing isolated direct effects, but in explaining how these constructs operate together as an integrated institutional capacity for DT in public administration.

In this study, digital governance capacity is understood as the institutional ability to align human skills, organizational change processes, and technology adoption mechanisms to produce meaningful digital transformation in public administration. This perspective goes beyond a technocentric view of digitalization. It assumes that public organizations do not transform only because technologies are available, but because they develop the capacity to mobilize digital competencies, manage organizational change, and embed digital tools into administrative routines. In this sense, the proposed framework contributes to the literature by explaining how digital competencies, change management, and technology adoption operate together as complementary mechanisms of public administration transformation.

In this scenario, it is necessary to advance toward integrated analytical models that allow understanding how individual capabilities, organizational practices, and contextual factors interact in state digital transformation processes. Analyzing these relationships from a structural perspective contributes to strengthening the debate on strategic digital governance and institutional modernization in emerging economies. Likewise, it makes it possible to generate empirical evidence that supports the design of public policies aimed at consolidating sustainable innovation processes in the public sector.

Therefore, the present study aims to examine the structural relationships among digital competencies, change management, technology adoption, and digital transformation in the public administration institutions in Lima, Peru, incorporating digital inclusion practices as a moderating variable. By integrating human, organizational, operational, and contextual dimensions into a single explanatory model, this research contributes to the literature on digital governance and public administration transformation.

The main theoretical contribution of this study is to conceptualize and empirically test digital governance capacity as an integrative framework for explaining digital transformation in public administration in an emerging economy context. While the individual relationships among digital competencies, change management, technology adoption, and digital transformation have been examined in prior research, this study reframes these relationships as part of an integrated institutional capacity mechanism. In this framework, digital competencies represent the human foundation, change management represents the organizational condition, technology adoption functions as the operational mechanism, and digital inclusion practices represent the inclusion-oriented dimension of digital governance.

This contribution is developed in three ways. First, the study integrates digital competencies, change management, technology adoption, and digital inclusion practices as complementary dimensions of institutional capacity rather than as isolated predictors. Second, it identifies technology adoption as the mechanism through which human capabilities and organizational change practices are translated into digital transformation outcomes. Third, it adapts the digital governance capacity perspective to the context of Lima, Peru, where public administration institutions operate under a common national digital transformation framework but differ in resources, mandates, and implementation capacity. In this way, the study contributes to digital governance theory by explaining digital transformation as an institutionally mediated process shaped by the alignment of human, organizational, technological, and inclusion-oriented conditions.

The remainder of this paper is structured as follows. Section 2 presents the theoretical framework of the study. Section 3 develops the research hypotheses. Section 4 describes the materials and methods applied in the research. Section 5 reports the empirical results. Section 6 discusses the findings and their implications for public administration and digital governance. Finally, Section 7 presents the main conclusions of the study.

2. Theoretical Framework

In the context of administrative modernization, digital transformation in the public sector has been widely recognized as a key mechanism for improving institutional efficiency and strengthening the relationship between the State and citizens. In this context, variables such as public servants’ digital competencies, organizational change management, and technology adoption emerge as critical factors that shape the effectiveness of digital initiatives. However, these dynamics are not uniform, as they are influenced by institutional efforts to promote digital inclusion, especially through training, equitable access strategies, awareness campaigns, and the adaptation of digital tools to users with different skill levels (Espina-Romero, 2025; Peixoto Rodriguez, 2025).

Although the literature converges in recognizing the importance of these variables, it still lacks integrative approaches that explain how they interact within public sector environments characterized by institutional rigidity and inequality. This fragmentation limits the explanatory capacity of existing models and justifies the need for more comprehensive approaches that integrate human, organizational, and contextual dimensions.

2.1. Digital Governance Capacity and Organizational Capacity in Public Administration Transformation

Digital governance capacity refers to the ability of public organizations to use digital technologies in a coordinated, strategic, and institutionally embedded manner. In public administration, this capacity is not limited to technological infrastructure. It also depends on civil servants’ skills, leadership support, change management practices, internal coordination, and the effective integration of technologies into administrative processes. Previous studies in the manuscript show that digital transformation requires more than technological investment, because bureaucratic inertia, limited readiness, and fragmented implementation can reduce the impact of digital initiatives (Norling, 2025; Santos, 2024; Tinjan, 2025).

From this perspective, organizational capacity is central to public administration transformation. Digital competencies represent the human capability required to work in digital environments. Change management represents the organizational capability to guide transition, reduce resistance, and support implementation. Technology adoption represents the practical mechanism through which digital tools become part of institutional routines. Digital transformation represents the institutional outcome of this alignment. Digital inclusion practices, in turn, reflect institutional efforts to address equity-related conditions that may affect the reach of digital governance. Therefore, the proposed framework explains digital transformation as the result of an integrated governance capacity rather than as the isolated effect of technology or individual skills.

2.2. Key Definitions and Conceptual Delimitation

(1)

Digital competencies (DCs): They are understood as the set of knowledge, skills, and attitudes necessary to effectively use digital technologies in the workplace. Recent studies highlight that DCs encompass from basic literacy to advanced skills in data management, cybersecurity, and digital collaboration (Cardoso & Gomes, 2025; Espina-Romero et al., 2024a, 2024b; Lopes et al., 2023; Medina, 2025). While Cardoso and Gomes (2025) and Lopes et al. (2023) emphasize the role of DCs as a direct enabler of digital performance, Medina (2025) suggests that their impact depends on their effective integration into organizational processes. This comparison indicates that DCs, although necessary, may not be sufficient to drive DT without complementary organizational mechanisms.

(2)

Change management (CM): It refers to the set of processes, strategies, and practices that facilitate organizational transition in response to the introduction of new technologies. The literature emphasizes the importance of organizational culture, internal communication, and leadership in overcoming resistance (Abawari et al., 2024; Barodi et al., 2024; Noroño Sánchez, 2025). While Abawari et al. (2024) and Barodi et al. (2024) focus on structural and cultural conditions for change, Noroño Sánchez (2025) highlights the role of leadership in aligning stakeholders. Taken together, these perspectives position CM not only as a support process but as a strategic capability that enables DT.

(3)

Technology adoption (TA): It involves the acceptance, use, and integration of digital tools into institutional processes. In the public sector, TA has been extensively explained through TAM and UTAUT, where performance expectancy, ease of use, and institutional trust are central determinants (Aristovnik et al., 2025; Sukma & Yamnill, 2025). While Aristovnik et al. (2025) emphasize behavioral intention, Sukma and Yamnill (2025) highlight contextual and institutional factors. This suggests that TA should be understood not only as an individual-level outcome but also as a process embedded in organizational dynamics.

(4)

Digital transformation (DT): Beyond the digitalization of processes, DT involves structural and cultural changes that redesign services and generate public value (Canonico et al., 2025; Ly, 2025; Santos, 2024). While Canonico et al. (2025) and Ly (2025) emphasize service redesign and digitalization, Santos (2024) frames DT as a capability reconfiguration process. This distinction reinforces the idea that DT is not merely technological but strategic and organizational.

(5)

Digital inclusion practices (DIP): These refer to institutional actions, strategies, and support mechanisms aimed at promoting equitable access to digital services, strengthening digital skills, raising awareness about digitalization, and adapting digital tools to users with different levels of technological readiness. Previous studies show that digital exclusion is associated with inequalities in access, skills, age, income, territory, and effective use of digital technologies (Cui et al., 2025; Espina, 2025; Singh & Chobotaru, 2022). In this study, DIP does not measure the digital divide itself, but rather institutional practices intended to reduce exclusion risks and support more inclusive digital transformation.

Taken together, these constructs form an integrated framework of digital governance capacity. DCs explain the individual readiness of public servants. CM explains the institutional support required to guide digital change. TA explains the actual integration of digital tools into public administration routines. DT captures the broader modernization of processes, services, and decision-making. DIP introduces the institutional inclusion dimension of digital governance. This integrated view clarifies that the contribution of the model lies not only in testing separate relationships, but in explaining how human, organizational, operational, and contextual conditions interact in public sector DT.

2.3. Review of Background and Previous Findings

Empirical evidence shows that DCs strengthen the public sector’s capacity to implement digital services, although significant gaps among public servants persist (Adewumi & Abasilim, 2024; Bilan et al., 2023). While Adewumi and Abasilim (2024) highlight improvements in service delivery, Bilan et al. (2023) point to persistent competency deficits, suggesting that the impact of DCs is uneven and context-dependent.

Similarly, CM has been identified as a decisive factor in facilitating the acceptance of new technologies and reducing resistance (Rehouma et al., 2020; Semenets-Orlova et al., 2023). Rehouma et al. (2020) emphasize structured change processes, whereas Semenets-Orlova et al. (2023) highlight leadership and communication. This comparison indicates that CM operates both as a formal process and as a relational capability.

Regarding TA, studies conducted in Europe, Asia, and Latin America converge in highlighting perceived usefulness, ease of use, and institutional trust as key determinants (Aranyossy, 2022; Crăciun et al., 2025). However, while Aranyossy (2022) emphasizes user perceptions, Crăciun et al. (2025) incorporate institutional dimensions, suggesting that TA is influenced by both individual and systemic factors.

DT has demonstrated positive effects on efficiency and service quality, but these outcomes are not guaranteed (Norling, 2025; Tinjan, 2025). While Norling (2025) highlights improvements in governance performance, Tinjan (2025) emphasizes the persistence of bureaucratic inertia. This divergence suggests that DT success depends on organizational alignment.

Finally, DIP emerges as a cross-cutting institutional response to digital exclusion risks. Studies in Canada, Hong Kong, and Vietnam show that socioeconomic, age, and gender disparities limit equitable access to digital services (Singh & Chobotaru, 2022; Tran Pham & Le Hoang Thuy To Nguyen, 2024; Wong & Ho, 2022). While Singh and Chobotaru (2022) focus on inequality, Tran Pham and Le Hoang Thuy To Nguyen (2024) highlight access gaps, and Wong and Ho (2022) emphasize demographic barriers. Together, these findings indicate that digital inclusion practices are important for improving the reach, equity, and social effectiveness of DT policies.

2.4. Gaps and Knowledge Voids

Despite the growing body of literature, important gaps remain. Most studies analyze DCs, CM, TA, and DT in isolation, limiting the development of integrative models. In addition, the mediating role of TA between DCs, CM, and DT has been insufficiently explored in Latin America. Furthermore, although digital inclusion is widely recognized as a policy priority, the moderating role of institutional digital inclusion practices has not been consistently tested in internal public administration models. This fragmentation reduces the explanatory capacity of existing approaches and limits the development of evidence-based digital governance strategies in public sector contexts.

2.5. Reference Models and Theories

This study is grounded in TAM and UTAUT to explain TA (Aristovnik et al., 2025; Sukma & Yamnill, 2025), and in Dynamic Capabilities Theory to explain DT (Santos, 2024). While TAM and UTAUT focus on individual acceptance, Dynamic Capabilities Theory emphasizes organizational adaptation, providing complementary perspectives. Additionally, CM is supported by the ADKAR model, which conceptualizes change as a sequence of awareness, desire, knowledge, ability, and reinforcement (Al-Alawi et al., 2019).

These theoretical perspectives are complementary. TAM and UTAUT explain why public servants and institutions accept and use digital technologies. ADKAR explains how public organizations can guide employees through digital change by creating awareness, desire, knowledge, ability, and reinforcement. Dynamic Capabilities Theory explains how public organizations reconfigure resources, routines, and capabilities to respond to changing environments. When combined, these perspectives support the idea that DT in public administration depends on both adoption behavior and organizational capacity. This theoretical integration strengthens the proposed framework by linking individual acceptance, change management, and capability reconfiguration within a digital governance approach.

2.6. Digital Transformation and Strategic Governance in the Public Administration

DT should be understood as a governance phenomenon that redefines interactions between the State and society (Norling, 2025; Tinjan, 2025). While Norling (2025) emphasizes performance improvements, Tinjan (2025) highlights structural limitations, indicating that DT requires more than technological investment. From a strategic perspective, digital governance integrates leadership, CM, and organizational capabilities. Aristovnik et al. (2025) show that structured adoption improves performance, while Santos (2024) emphasizes capability reconfiguration. This suggests that TA functions as a mechanism that connects capabilities with transformation outcomes.

Furthermore, the interaction between DCs, CM, and TA represents a shift toward more adaptive public administration models in Latin America. In contexts such as Lima, these factors are critical to overcoming institutional rigidity. Finally, incorporating DIP introduces an equity dimension. Singh and Chobotaru (2022) warn about inequality risks, while Tran Pham and Le Hoang Thuy To Nguyen (2024) highlight access limitations. This confirms that DT has social and political implications beyond organizational performance.

Therefore, the proposed model should be understood as a digital governance capacity framework. It explains how public institutions convert individual digital skills and organized change practices into transformation outcomes through the effective adoption of technologies. This approach is especially relevant in emerging public administrations, where digital reforms often face institutional rigidity, uneven readiness, and social inequality. The framework also helps clarify why DT requires the alignment of human competencies, organizational support, technology use, and inclusive governance conditions.

2.7. Relationship with the Research Hypotheses

The literature supports the proposed hypotheses. DCs and CM influence TA (Adewumi & Abasilim, 2024; Bilan et al., 2023; Rehouma et al., 2020; Semenets-Orlova et al., 2023), supporting H1 and H2. Direct effects on DT are supported by Canonico et al. (2025) and Santos (2024), supporting H3 and H4. The mediating role of TA is supported by Aristovnik et al. (2025) and Sukma and Yamnill (2025), supporting H5–H7. Finally, the moderating role of DIP is supported by Odei-Appiah and Owusu (2023) and Tran Pham and Le Hoang Thuy To Nguyen (2024), supporting H8.

2.8. Final Synthesis

In summary, DT in the public sector results from the interaction between DCs, CM, TA, and DIP. DCs provide the human foundation for digital work. CM creates the organizational conditions for transition. TA operates as the mechanism that converts capabilities and support into concrete use of technologies. DIP represents a contextual condition related to equity and inclusion. This framework contributes to public administration literature by explaining DT as an institutionally mediated process. It also extends digital governance research by showing that transformation depends on the capacity of public organizations to align people, processes, technologies, and contextual conditions.

3. Development of Hypotheses

3.1. Digital Competencies and Technology Adoption

DCs are essential for the effective use of digital technologies in public administration. According to the Technology Acceptance Model (TAM) proposed by Davis (1989), individuals are more likely to adopt technologies when they perceive them as useful and easy to use. Similarly, the Unified Theory of Acceptance and Use of Technology (UTAUT) developed by Venkatesh et al. (2003) suggests that users are more willing to adopt technologies when they possess the knowledge and skills required to use them effectively.

In public organizations, civil servants with stronger DCs are better prepared to understand, operate, and integrate digital tools into their daily activities. These competencies reduce uncertainty, increase confidence, and facilitate the use of digital systems. Previous studies have shown that digital skills positively influence technology adoption in both public and private organizations (Ly, 2025; van Laar et al., 2017). Therefore, DCs are expected to positively influence TA in public administration.

H1. 

DCs positively influence TA.

3.2. Change Management and Technology Adoption

TA in public organizations depends not only on individual capabilities but also on organizational support. CM helps employees adapt to new technologies by reducing resistance and facilitating organizational transitions. The ADKAR model proposed by Hiatt (2006) explains that successful change requires awareness, desire, knowledge, ability, and reinforcement.

In public administration, digital initiatives often face resistance due to established routines, bureaucratic structures, and uncertainty about new processes. Effective CM creates favorable conditions for employees to accept and use digital technologies. Previous studies have found that CM contributes significantly to successful technology implementation and adoption (AlNuaimi et al., 2022; Kotter, 1996). Therefore, CM is expected to positively influence TA.

H2. 

CM positively influences TA.

3.3. Digital Competencies and Digital Transformation

DT requires employees who can effectively use digital technologies and adapt to changing work environments. DCs provide the knowledge and skills necessary to support organizational modernization and innovation. From the perspective of Dynamic Capabilities Theory (Teece et al., 1997), organizations need human capabilities to identify opportunities, adapt resources, and respond to environmental changes. In public administration, employees with strong DCs can contribute to process improvement, service innovation, and data-driven decision-making. Previous research has highlighted the importance of digital skills as a driver of DT (Ly, 2025; Sousa & Rocha, 2019; Verhoef et al., 2021). Therefore, DCs are expected to positively influence DT.

H3. 

DCs positively influence DT.

3.4. Change Management and Digital Transformation

CM plays a critical role in enabling DT because organizational transformation requires changes in structures, processes, and employee behaviors. Digital initiatives often fail when organizations focus only on technology and neglect the human and organizational dimensions of change.

According to Kotter (1996) and Hiatt (2006), successful transformation depends on leadership support, communication, employee engagement, and reinforcement mechanisms. In public organizations, CM helps align employees with transformation goals and facilitates the implementation of new digital practices. Previous studies have reported positive relationships between CM and DT outcomes (AlNuaimi et al., 2022). Therefore, CM is expected to positively influence DT.

H4. 

CM positively influences DT.

3.5. Technology Adoption and Digital Transformation

TA is a central mechanism through which organizations achieve DT. While digital technologies may be available, transformation occurs only when these technologies are actively used and integrated into organizational processes. TAM (Davis, 1989) and UTAUT (Venkatesh et al., 2003) explain how technology use emerges from user acceptance and behavioral intentions. In public administration, the adoption of digital tools can improve efficiency, service delivery, communication, and decision-making. Previous studies have consistently identified TA as a key driver of DT (AlNuaimi et al., 2022; Verhoef et al., 2021). Therefore, TA is expected to positively influence DT.

H5. 

TA positively influences DT.

3.6. Mediation of Technology Adoption

Although DCs and CM are important organizational resources, they do not automatically generate DT. Their impact becomes effective when employees actually adopt and use digital technologies. Dynamic Capabilities Theory (Teece et al., 1997) suggests that organizational resources create value when they are mobilized and integrated into operational activities.

In this context, TA acts as the mechanism that converts employee competencies and organizational support into transformation outcomes. Employees with strong DCs are more likely to adopt digital technologies, and organizations with effective CM create favorable conditions for technology use. Through TA, these factors contribute to process modernization and organizational transformation.

Previous studies have suggested that technology use often mediates the relationship between organizational capabilities and transformation outcomes (Ly, 2025; Venkatesh et al., 2003; Verhoef et al., 2021). Therefore, TA is expected to mediate the relationships between DCs and DT, and between CM and DT.

H6. 

TA mediates the relationship between DCs and DT.

H7. 

TA mediates the relationship between CM and DT.

3.7. Moderation of the Digital Inclusion Practices

DIP refer to organizational efforts that promote equitable access to digital technologies, digital training, awareness, and support for users with different levels of digital readiness. In public administration, digital transformation should not only improve internal efficiency but also ensure that digital initiatives are accessible and usable for diverse groups.

The public value perspective emphasizes that digital governance should promote inclusion and equal opportunities for participation and access to services (Cordella & Bonina, 2012; Criado & Guevara-Gómez, 2021). When organizations implement strong DIP, employees and users may be better prepared to benefit from adopted technologies. Therefore, the positive effect of TA on DT may become stronger when DIP are more developed. In contrast, limited inclusion practices may reduce the transformational benefits of technology adoption.

H8. 

DIP positively moderate the relationship between TA and DT, such that the relationship is stronger when DIP are higher.

Once the Hypotheses Development section is completed, Figure 1 below shows the proposed model for this study.

4. Materials and Methods

4.1. Research Design

This study followed a quantitative, non-experimental, cross-sectional design. The objective was to examine the structural relationships among DCs, CM, TA, DT, and the moderating role of DIP in public administration institutions in Lima, Peru. A cross-sectional design is appropriate when the goal is to analyze relationships among variables at a single point in time (Creswell & Creswell, 2018). Because the study does not manipulate variables but rather examines naturally occurring organizational perceptions, it is classified as non-experimental (Kerlinger & Howard, 2002).

The analytical approach was based on Partial Least Squares Structural Equation Modeling (PLS-SEM), which is recommended when the research aims to predict key constructs and assess complex mediation and moderation relationships (Hair et al., 2021; Henseler et al., 2015). PLS-SEM is particularly suitable for exploratory and prediction-oriented models in social sciences and public administration contexts.

4.2. Population and Sample

The population consisted of public sector employees working in public administration institutions located in Lima, Peru. For the purposes of this study, three levels of government were considered. The national level refers to employees working in ministries, national agencies, and autonomous public bodies whose headquarters or administrative offices are located in Lima. The regional level refers to employees working in regional government units or related administrative bodies with jurisdiction in the Lima region. The local level refers to employees working in municipal governments, including provincial and district municipalities located in Metropolitan Lima and the broader Lima region.

Therefore, the empirical scope of this study is limited to public administration institutions located in Lima, Peru. The study does not include broader public-sector entities such as state-owned enterprises, public universities, public hospitals, or other publicly funded organizations. In this study, the term public servants is used as an umbrella category to refer to individuals who perform administrative, professional, technical, or managerial functions within public administration institutions. This category included career civil servants, appointed personnel, and contractual staff working in public organizations at the time of data collection.

Temporary workers and external consultants were not intentionally targeted unless they performed formal institutional duties within the participating public organizations. However, employment modality was not used as a stratification criterion, and the survey did not collect sufficiently detailed information to compare respondents by type of employment contract. Therefore, the results should be interpreted as reflecting the perceptions of public sector employees broadly, rather than the perceptions of a specific employment category.

Data were collected through an online questionnaire distributed to public servants across ministries, autonomous agencies, municipalities, and regional government units. A total of 358 valid responses were obtained. This sample size exceeds the minimum requirements suggested for PLS-SEM analysis. According to Hair et al. (2021), the minimum sample size should satisfy the “10-times rule” or preferably be determined based on statistical power analysis. With multiple predictors in the model, the sample size of 358 provides adequate statistical power for detecting medium and small effects (Cohen, 2013).

The demographic profile of respondents is presented in

Table 1. Demographics Profile.

Variable	Category	n	%
Gender	Female	167	46.60
	Male	191	53.40
Age (years)	22–29	68	19.00
	30–39	157	43.90
	40–49	85	23.70
	50–59	36	10.10
	≥60	12	3.40
Educational Attainment	Technical Degree	11	3.10
	Bachelor’s Degree	58	16.20
	Professional Degree (Licensed)	230	64.20
	Specialist Certification	15	4.20
	Master’s Degree	40	11.20
	Doctoral Degree (PhD)	4	1.10
Level of Government	National	209	58.40
	Regional	61	17.00
	Local	88	24.60

. The sample included participants from national (58.4%), regional (17.0%), and local government (24.6%) levels, with balanced gender distribution and diverse educational attainment.

4.3. Measurement Instrument

The measurement instrument was adapted from the validated multivariable model developed by (Espina-Romero, 2025), which originally included 11 constructs. For the present study, five constructs were selected based on theoretical alignment with digital governance and public administration transformation. Each construct was measured using four reflective indicators, resulting in a total of 20 items. All items were assessed using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), consistent with standard measurement practices in organizational research (Joshi et al., 2015).

The construct measurement was reviewed to ensure stronger alignment with the public administration literature. Each construct was operationalized according to its role in public sector digital governance. DCs were measured as civil servants’ ability to use digital tools, communicate through digital platforms, participate in digital training, and work with collaboration systems. CM was measured through institutional training, planning, internal communication, and employee support during technological change. TA was measured through institutional promotion of technologies, employee openness, usability, and improvement of administrative processes. DT was measured through efficiency, user interaction, decision-making optimization, and document digitization. DIP was measured through training for vulnerable groups, equitable access strategies, awareness campaigns, and adaptation of digital tools to different skill levels.

The full list of questionnaire items is presented in

Table 2. Questionnaire applied to the present study.

Variable	Item	Questionnaire Item	Supporting Authors	Public Administration Rationale
Digital Competencies	DCs1	“I have advanced skills for using digital tools in my daily work”	(Cardoso & Gomes, 2025; Lopes et al., 2023)	Measures civil servants’ digital readiness for public service work.
	DCs2	“I feel comfortable using digital platforms to communicate with colleagues and citizens”	(Cardoso & Gomes, 2025)
	DCs3	“I actively participate in training to improve my digital skills”	(Cordella et al., 2024; Lopes et al., 2023)
	DCs4	“I have experience using digital collaboration systems (such as Microsoft Teams, Google Workspace, etc.)”	(Cardoso & Gomes, 2025)
Change Management	CM1	“My institution provides adequate training for the adoption of new technologies”	(Barodi et al., 2024)	Measures institutional support for managing digital reform.
	CM2	“There is a clear plan to manage digital change in my institution”	(Rehouma et al., 2020)
	CM3	“Internal communication is effective during technological change processes”	(Barodi et al., 2024)
	CM4	“Employees receive support during the adoption of new digital platforms”	(Semenets-Orlova et al., 2023)
Technology Adoption	TA1	“My institution actively promotes the use of new technologies to improve public services”	(Sukma & Yamnill, 2025)	Measures acceptance and integration of technologies into administrative routines.
	TA2	“Staff in my organization are open to using new digital tools”	(Aristovnik et al., 2025)
	TA3	“The technological tools implemented in my organization are easy to use and efficient”	(Hameduddin et al., 2025)
	TA4	“The use of digital technologies has significantly improved administrative processes”	(Engkus, 2025)
Digital Transformation	DT1	“Digital transformation has improved the efficiency of administrative processes”	(Ly, 2025; Santos, 2024)	Measures modernization of processes, services, and decision-making.
	DT2	“My institution has implemented digital platforms that facilitate user interaction”	(Canonico et al., 2025)
	DT3	“Digital transformation has optimized decision-making in my organization”	(Tinjan, 2025)
	DT4	“Process digitization has significantly reduced the use of physical documents”	(Hamdy et al., 2025)
Digital Inclusion Practices	DIP1	“My organization develops technology training programs for vulnerable populations”	(Boksova et al., 2021)	Measures inclusive digital governance and equitable access conditions.
	DIP2	“Strategies have been implemented to ensure equitable access to digital services”	(Singh & Chobotaru, 2022)
	DIP3	“Awareness campaigns on the importance of digitization have been carried out”	(Cui et al., 2025)
	DIP4	“Digital tools are adapted to people with different levels of technological skills”	(Cui et al., 2025)

of the manuscript.

These indicators were selected because they reflect key dimensions of public administration. The DCs items capture the digital readiness of civil servants. The CM items capture institutional mechanisms used to manage reform and reduce resistance during digital change. The TA items capture the organizational use and acceptance of technologies in administrative routines. The DT items capture concrete modernization outcomes related to efficiency, service interaction, decision-making, and paperless administration. The DIP items capture the inclusive dimension of digital governance by focusing on access, training, awareness, and adaptation for users with different digital skill levels. Therefore, the measurement model is grounded not only in general organizational research, but also in the specific requirements of digital transformation in public administration.

Because the instrument had been previously validated, construct operationalization followed established procedures. However, reliability and validity were re-evaluated in the present sample to ensure measurement robustness in the public sector context (Hair et al., 2021).

4.4. Data Collection Procedure

Data were collected through a structured online survey administered to public servants between January and May 2025. The questionnaire was distributed online to employees working in ministries, autonomous agencies, municipalities, and regional government units located in Lima, Peru. Participation was voluntary, and confidentiality was guaranteed. Respondents were informed about the academic purpose of the study. Cross-sectional survey data are appropriate for examining perceptions and organizational attitudes in public administration contexts. To reduce common method bias, items were clearly worded, constructs were conceptually separated, and anonymity was ensured (Podsakoff et al., 2003).

4.5. Data Analysis Strategy

The data analysis followed a two-step approach consistent with PLS-SEM guidelines (Hair et al., 2021):

4.5.1. Step 1: Assessment of the Measurement Model

The reflective measurement model was evaluated using:

(a)

Indicator loadings (≥0.70 recommended).

(b)

Internal consistency reliability (Cronbach’s Alpha, Composite Reliability).

(c)

Convergent validity (Average Variance Extracted ≥ 0.50).

(d)

Discriminant validity (Fornell–Larcker criterion and HTMT ratio).

Cronbach’s Alpha was assessed following Nunnally and Bernstein (1994). Convergent validity was evaluated using the AVE threshold proposed by Fornell and Larcker (1981). Discriminant validity was assessed using the HTMT criterion recommended by Henseler et al. (2015).

4.5.2. Step 2: Assessment of the Structural Model

The structural model was evaluated using:

(a)

Path coefficients (β).

(b)

Bootstrapping with resampling to test statistical significance.

(c)

Confidence intervals (95%).

(d)

Coefficient of determination (R²).

(e)

Effect sizes (f²).

(f)

Mediation analysis using Variance Accounted For (VAF).

(g)

Moderation analysis using product-indicator interaction.

Bootstrapping was used to assess the significance of direct, indirect, and moderating effects (Hair et al., 2021). Effect sizes were interpreted using Cohen’s (2013) thresholds. Mediation effects were classified according to Nitzl et al. (2016). Moderation was tested using interaction modeling procedures recommended by Henseler and Fassott (2010) and Chin et al. (2003). All analyses were conducted using SmartPLS 4 software (v.4.1.1.6) (Ringle et al., 2024).

4.6. Ethical Considerations

This study followed the ethical principles outlined in the Declaration of Helsinki (1975, revised 2013). Participation was voluntary, anonymous, and based on informed consent. No sensitive personal data were collected. Given the minimal risk nature of the study and its non-interventional design, formal ethical approval was not required.

4.7. Declaration on the Use of AI-Assisted Technologies

During the development and processing of this study, various technological tools were employed as support instruments. These included Microsoft Word (Microsoft 365 version) for grammar and stylistic revision, Microsoft Excel (Microsoft 365 version) for statistical data analysis, ChatGPT (GPT-5.3) and DeepL (web version) for translation verification and enhancement of textual clarity, and Google Scholar (web version) for the identification and validation of academic sources. It is important to note that the use of these tools did not substitute the authors’ analytical reasoning or scientific judgment. The following section presents the empirical results obtained from the PLS-SEM analysis, including the evaluation of the measurement model and the structural relationships proposed in the research framework.

5. Results

5.1. Reliability and Validity Analysis of the Measurement Model

Table 3. Model reliability and validity.

Construct	Item	Loadings	Cronbach’s Alpha	rho_a	rho_c	AVE
Change Management (CM)	CM1	0.875	0.893	0.893	0.926	0.758
	CM2	0.900
	CM3	0.877
	CM4	0.828
Digital Competencies (DCs)	DCs1	0.866	0.845	0.855	0.895	0.682
	DCs2	0.773
	DCs3	0.843
	DCs4	0.817
Digital Inclusion Practices (DIP)	DIP1	0.905	0.890	0.900	0.924	0.752
	DIP2	0.903
	DIP3	0.853
	DIP4	0.804
Digital Transformation (DT)	DT1	0.814	0.858	0.868	0.904	0.701
	DT2	0.838
	DT3	0.891
	DT4	0.803
Technology Adoption (TA)	TA1	0.813	0.726	0.740	0.829	0.549
	TA2	0.732
	TA3	0.669
	TA4	0.745
Moderation of DIP in TA → DT	DIP × TA	1.000

presents the indicators of reliability and convergent validity of the constructs of the model. First, the factor loadings of most items exceed the recommended value of 0.700 (Hair et al., 2021), with the exception of one item in TA (TA3 = 0.669). Although this value is below the ideal threshold, it is considered acceptable in exploratory studies (Hair et al., 2021).

Regarding internal consistency, all constructs show Cronbach’s Alpha values above 0.700, which indicates good reliability (Nunnally & Bernstein, 1994). The results obtained are: CM (0.893), DCs (0.845), DIP (0.890), DT (0.858), and TA (0.726). These values are consistent with the rho_a and rho_c (composite reliability) indices, which exceed the recommended threshold of 0.700 in all cases (Henseler et al., 2015). Regarding the Average Variance Extracted (AVE), all constructs meet the minimum criterion of 0.500 proposed by Fornell and Larcker (1981), which confirms convergent validity: CM = 0.758, DCs = 0.682, DIP = 0.752, DT = 0.701, and TA = 0.549. Although the lowest value corresponds to TA, it remains within acceptable limits, implying that more than 50% of the variance of the items is explained by the construct.

Additionally, the table includes the moderating effect DIP × TA → DT, whose loading = 1.000 corresponds to the statistical construction of the interaction. In this type of constructs formed by a single indicator resulting from the product of the variables involved, parameters such as Cronbach’s Alpha or AVE are not evaluated (Chin et al., 2003; Cronbach, 1951). Their validity is later verified in the structural model by assessing the significance of the moderating path (Henseler & Fassott, 2010). The inclusion of this interaction is consistent with the hypothesis that stronger DIP intensifies the effect of TA on DT. These results confirm that the constructs present adequate reliability and convergent validity, allowing progression to the analysis of discriminant validity and the evaluation of the structural model.

5.2. Discriminant Validity Analysis: Fornell–Larcker Criterion

Table 4. Discriminant validity matrix according to the Fornell and Larcker criterion.

	CM	DCs	DIP	DT	TA
CM	0.871
DCs	0.468	0.826
DIP	0.555	0.252	0.867
DT	0.641	0.497	0.425	0.837
TA	0.734	0.542	0.544	0.671	0.741

presents the results of the Fornell and Larcker criterion to evaluate the discriminant validity of the model. According to this approach, the square root of the AVE of each construct (shown on the diagonal) must be greater than the correlations with the other constructs in the matrix (Fornell & Larcker, 1981).

The obtained values meet this criterion:

(a)

CM (0.871) is higher than its correlations with the other constructs (max. 0.734 with TA).

(b)

DCs (0.826) exceeds its correlations with the rest (max. 0.542 with TA).

(c)

DIP (0.867) is higher than all its correlations (max. 0.555 with CM).

(d)

DT (0.837) is above its correlations (max. 0.671 with TA).

(e)

TA (0.741) is also higher than its correlations (max. 0.734 with CM).

These results indicate that each construct shares more variance with its own indicators than with those of other constructs, which confirms the discriminant validity of the model (Hair et al., 2021). The Fornell–Larcker matrix confirms that the constructs of the model are empirically distinct and measure different concepts, ensuring the discriminant validity required to proceed with the structural analysis.

5.3. Discriminant Validity Analysis: HTMT Criterion

Table 5. Discriminant validity matrix based on the HTMT criterion.

	CM	DCs	DIP	DT	TA	DIP × TA
CM
DCs	0.530
DIP	0.623	0.268
DT	0.720	0.579	0.474
TA	0.899	0.668	0.670	0.842
DIP × TA	0.112	0.123	0.188	0.105	0.256

presents the values of the heterotrait–monotrait ratio (HTMT), which evaluates discriminant validity among the constructs of the model. According to Henseler et al. (2015), HTMT values must be below 0.850 (more stringent criterion) or 0.900 (more lenient criterion) to confirm that the constructs are empirically distinct.

The results show that most relationships meet the criterion. However, it is observed that the relationship between CM and TA reaches a value of 0.899, which is at the upper limit of the 0.90 threshold. This indicates that both constructs are highly correlated, although still within the acceptable range. The other relationships show lower values: CM–DT (0.720), CM–DIP (0.623), DCs–TA (0.668), DIP–TA (0.670), and DT–TA (0.842), all below 0.900, confirming their discriminant validity. Finally, the moderating construct DIP × TA shows very low values with all others (max. 0.256), indicating the absence of collinearity problems and confirming that the interaction is clearly differentiated from the main constructs. Overall, the HTMT results reinforce the evidence of discriminant validity already observed with the Fornell–Larcker criterion, ensuring that the constructs of the model measure distinct concepts and do not excessively overlap.

5.4. Direct, Mediating, and Moderating Path Assessment

Table 6. Results of direct paths, indirect paths, and moderating relationship in the structural model.

	Sample (O)	t-Value	p-Value	2.5% Confidence Interval	97.5% Confidence Interval	R2	Path Interpretation
H1: DCs → TA	0.254	5.983	0.000	0.169	0.335	58.80%	Significant positive path
H2: CM → TA	0.615	16.275	0.000	0.537	0.685	61.50%	Strong significant positive path
H3: DCs → DT	0.160	2.465	0.014	0.038	0.294	51.70%	Weak significant positive path
H4: CM → DT	0.283	4.323	0.000	0.157	0.410	51.70%	Significant positive path
H5: TA → DT	0.362	4.897	0.000	0.216	0.503	51.70%	Significant positive path
H6: DCs → TA → DT	0.092	3.864	0.000	0.051	0.145		Low partial mediation (VAF = 10.09%)
H7: CM → TA → DT	0.223	4.663	0.000	0.131	0.318		Moderate partial mediation (VAF = 32.47%)
H8: DIP × TA → DT	0.016	0.602	0.547	−0.032	0.070		Non-significant moderating effect

presents the results of the structural model, including the direct paths, indirect paths, and moderating relationship. The values of the sample (Sample O), t-statistics, p-values, 95% confidence intervals, coefficients of determination (R²), and the VAF values used to assess the indirect relationships are included.

5.4.1. Direct Paths

The direct paths were statistically significant in almost all hypotheses:

• H1 (DCs → TA, β = 0.254, t = 5.983, p < 0.001): Significant positive path.
• H2 (CM → TA, β = 0.615, t = 16.275, p < 0.001): strong and significant positive path.
• H3 (DCs → DT, β = 0.160, t = 2.465, p = 0.014): positive path although of smaller magnitude.
• H4 (CM → DT, β = 0.283, t = 4.323, p < 0.001): Significant positive path.
• H5 (TA → DT, β = 0.362, t = 4.897, p < 0.001): significant positive path.

The R² values show that the explained variance is moderate to high: TA = 58.8% and 61.5%, while DT = 51.7%, indicating a model with adequate explanatory capacity (Hair et al., 2021).

5.4.2. Indirect Paths and Mediation Assessment

The indirect paths were evaluated using VAF (Variance Accounted For), which is calculated with the formula:

$$\mathrm{VAF}={\displaystyle \frac{\mathrm{a}\ast \mathrm{b}}{\mathrm{a}\ast \mathrm{b}+{\mathrm{c}}^{\prime }}}{\displaystyle \frac{\mathrm{Indirect} \mathrm{effect}}{\mathrm{Total} \mathrm{effect}}}$$

In this calculation, the indirect path is the product of the partial paths, and the total path is the sum of the direct path and the indirect path (Hair et al., 2021).

• H6 (DCs → TA → DT, β = 0.092, t = 3.864, p < 0.001, VAF = 10.09%): corresponds to a very low partial mediation, since VAF < 20% (Nitzl et al., 2016).
• H7 (CM → TA → DT, β = 0.223, t = 4.663, p < 0.001, VAF = 32.47%): considered a moderate partial mediation, as VAF is between 20% and 80%.

5.4.3. Moderating Relationship

• H8 (DIP × TA → DT, β = 0.016, t = 0.602, p = 0.547): it is not significant, since p-value > 0.05 and the confidence interval includes zero. Therefore, the moderating relationship of DIP in the TA → DT path was not statistically supported.

5.5. Effect Size Test

Table 7. Summary of effect size test (f²).

	f2	t-Value	p-Value	2.5% Confidence Interval	97.5% Confidence Interval
DCs → TA	0.122	2.831	0.005	0.121	0.121
CM → TA	0.716	5.571	0.000	0.747	0.739
DCs → DT	0.036	1.112	0.266	−0.043	0.037
CM → DT	0.069	1.986	0.047	0.022	0.022
TA → DT	0.100	2.394	0.017	0.094	0.094
DIP × TA → DT	0.001	0.218	0.827	−0.076	0.039

presents the f² values, which allow assessing the relative contribution of each predictor to the explained variance of the endogenous constructs. According to Cohen (2013), f² values of 0.020 indicate a small effect, 0.150 a medium effect, and 0.350 a large effect.

• DCs → TA (f² = 0.122, p = 0.005): small but statistically significant f² value.
• CM → TA (f² = 0.716, p < 0.001): very large and statistically significant f² value, indicating a strong contribution of CM to the explained variance of TA.
• DCs → DT (f² = 0.036, p = 0.266): small effect size; despite being statistically significant in the structural model, its relative contribution to the explained variance of DT is weak.
• CM → DT (f² = 0.069, p = 0.047): small but significant effect.
• TA → DT (f² = 0.100, p = 0.017): small-to-moderate and significant effect, indicating that TA shows a significant positive relationship with DT.
• DIP × TA → DT (f² = 0.001, p = 0.827): practically null and non-significant effect, indicating that the moderating relationship involving DIP was not statistically supported.

These results show that the largest f² value in the model corresponds to the CM → TA relationship, while the other relationships show smaller magnitudes. Figure 2 shows the final measurement model generated with SmartPLS 4.

Table 8. Summary of hypothesis assessment.

Hypothesis/Relationship	Type of Relationship	Confirmation
H1: DCs → TA	Direct path	Supported
H2: CM → TA	Direct path	Supported
H3: DCs → DT	Direct path	Supported (positive but weak path)
H4: CM → DT	Direct path	Supported
H5: TA → DT	Direct path	Supported
H6: DCs → TA → DT	Indirect path/Mediation	Supported (partial)
H7: CM → TA → DT	Indirect path/Mediation	Supported (partial)
H8: DIP × TA → DT	Moderating relationship	Not supported

then presents a summary of the confirmation of the hypotheses formulated, based on the results obtained using the structural equation model.

6. Discussion

6.1. Main Findings

This study examined the structural relationships among DCs, CM, TA, DT, and DIP in public administration institutions in Lima. Because the research design was cross-sectional and perception-based, the findings should be interpreted as structural associations rather than causal effects. The results show that DCs and CM are positively associated with TA. CM and TA also show significant positive relationships with DT. In addition, TA mediates the relationships between DCs and DT, and between CM and DT. However, DIP did not significantly strengthen the relationship between TA and DT.

The interpretation of H3 requires specific attention. The relationship between DCs and DT was statistically significant (β = 0.160, p = 0.014), so H3 was supported. However, the effect size was small (f² = 0.036). This means that DCs are associated with DT, but their explanatory relevance is limited when they are considered alone. This result suggests that digital skills become more relevant when they are connected to institutional support, organizational routines, and TA processes.

The strongest relationship in the model was observed between TA and DT. This finding suggests that the institutional use of technologies may be a central mechanism associated with DT in public administration. Individual skills are important, but they do not appear to be sufficient by themselves. They gain greater relevance when they are translated into concrete adoption processes and embedded in administrative routines (Aristovnik et al., 2025; Santos, 2024).

6.2. Theoretical Discussion and Comparison with Prior Studies

The findings of this study contribute to the theoretical discussion in three main ways. First, they extend technology adoption theories by showing that TA in public administration should not be interpreted only as an individual acceptance process. TAM and UTAUT explain technology adoption mainly through user perceptions, behavioral intention, usefulness, ease of use, and facilitating conditions. However, in public administration institutions, TA is also shaped by institutional conditions, organizational routines, and the capacity of public organizations to support employees during digital change. In this sense, the results are consistent with Aristovnik et al. (2025) and Sukma and Yamnill (2025), who emphasize the relevance of technology acceptance and use for the modernization of public administration. However, this study extends that discussion by showing that TA becomes more theoretically meaningful when it is examined together with DCs and CM. Therefore, TA should not be viewed only as the acceptance of digital tools, but as the operational mechanism through which public administration institutions embed technologies into administrative routines.

Second, the findings refine the role of DCs in DT in public administration. Previous studies have emphasized that digital skills among civil servants are important for improving digital readiness, service delivery, and administrative performance (Adewumi & Abasilim, 2024; Bilan et al., 2023; Cardoso & Gomes, 2025; Lopes et al., 2023). The present study does not contradict this view. However, it suggests that DCs should not be interpreted as an isolated driver of DT. Although DCs are positively associated with DT, their explanatory relevance is stronger when they are connected to TA and supported by CM. This finding refines the digital competencies literature by showing that digital skills become more relevant when they are institutionally mobilized, translated into actual technology use, and connected to organizational transformation processes within public administration institutions.

Third, the findings refine Dynamic Capabilities Theory in the public administration context. From this perspective, DT does not depend only on the possession of resources, but on the capacity of institutions to mobilize, organize, and reconfigure those resources in response to changing demands. This interpretation is consistent with Santos (2024), who argues that dynamic capabilities are relevant for understanding DT in public administration. In the proposed model, DCs represent the human capability base, CM represents the organizational condition that supports transition, and TA represents the operational mechanism that connects internal capabilities with DT. Therefore, the study contributes to Dynamic Capabilities Theory by showing that public administration institutions need more than digital resources or skilled employees. They also need mechanisms that convert those resources into new administrative routines and modernization outcomes.

The findings related to CM also strengthen this interpretation. Prior studies have shown that leadership, communication, training, participation, and organizational culture are critical for change processes in public administration (Abawari et al., 2024; Barodi et al., 2024; Rehouma et al., 2020; Semenets-Orlova et al., 2023). The present study supports this perspective, but it also adds that CM is not only a support activity for digital reform. It is part of Digital Governance Capacity because it creates the organizational conditions that allow technologies to be accepted, used, and integrated into institutional processes. In bureaucratic public administration institutions, where routines, hierarchy, and resistance may slow down digital initiatives, CM becomes a necessary condition for connecting digital strategies with actual transformation.

The non-significant moderating role of DIP provides an additional theoretical nuance. Prior studies have shown that digital inequality can affect access to online government services, citizen engagement, and the public value generated by e-government initiatives (Singh & Chobotaru, 2022; Tran Pham & Le Hoang Thuy To Nguyen, 2024; Wong & Ho, 2022). However, those studies mainly emphasize citizen-facing relationships. In contrast, the present study focuses on internal public administration processes among public servants. This difference may explain why DIP did not significantly moderate the relationship between TA and DT. The finding does not imply that DIP is irrelevant. Rather, it suggests that DIP may operate more strongly in models focused on citizens, service users, public value, or digital access, while internal DT in public administration may depend more directly on CM, DCs, and TA.

Overall, these findings clarify the theoretical contribution of the study. The model does not claim that the individual relationships among DCs, CM, TA, DT, and DIP are entirely new. Instead, its contribution lies in reframing these relationships as part of an integrated Digital Governance Capacity mechanism. This framework explains DT as an institutionally mediated process in which human capabilities, organizational support, technology use, and inclusion-oriented practices interact within public administration. Therefore, the study extends technology adoption theory, refines Dynamic Capabilities Theory in the public administration context, and clarifies the possible level of operation of DIP in internal digital transformation models.

6.3. Theoretical Contribution: Digital Governance Capacity as an Integrative Framework

The main theoretical contribution of this study lies in conceptualizing and empirically testing Digital Governance Capacity as an integrative framework for explaining DT in public administration. This study does not claim that DCs, CM, TA, DT, or DIP are new constructs by themselves. These constructs have already been examined in previous studies on digital transformation, technology adoption, digital competencies, change management, and digital inclusion. Rather, the contribution of this study lies in the way these constructs are theoretically articulated as complementary components of an institutional capacity mechanism.

In this framework, Digital Governance Capacity is understood as the institutional ability to align human capabilities, organizational change conditions, technology-use mechanisms, and inclusion-oriented practices to support DT. DCs represent the human foundation of this capacity because public servants need digital skills to understand, use, and adapt digital tools in their daily work. This interpretation is consistent with Adewumi and Abasilim (2024), Bilan et al. (2023), Cardoso and Gomes (2025), and Lopes et al. (2023), who highlight the relevance of digital skills for the digitalization of public administration. However, the present study refines this view by showing that DCs should not be understood as an isolated driver of DT. Their theoretical relevance increases when they are connected to CM and translated into actual TA.

CM represents the organizational condition of Digital Governance Capacity. In public administration, digital transformation requires more than individual skills or technological infrastructure. It also requires leadership, communication, planning, participation, support, and resistance management. This argument is consistent with Abawari et al. (2024), Barodi et al. (2024), Rehouma et al. (2020), and Semenets-Orlova et al. (2023), who emphasize the role of change management in public administration reform. The present study extends this perspective by positioning CM not only as a managerial support practice, but as a core component of digital governance capacity. In bureaucratic public administration institutions, CM helps convert digital reform from a formal institutional agenda into concrete organizational practices.

TA represents the operational mechanism of Digital Governance Capacity. Technology adoption theories, especially TAM and UTAUT, explain how users accept and use digital technologies based on usefulness, ease of use, behavioral intention, facilitating conditions, and institutional support (Davis, 1989; Venkatesh et al., 2003). Previous studies have also emphasized the importance of TA in the modernization of public administration and digital government development (Aristovnik et al., 2025; Sukma & Yamnill, 2025). This study contributes to that literature by showing that TA operates as the mechanism through which DCs and CM become associated with DT. In other words, public administration institutions may have digitally competent employees and change-oriented practices, but these resources generate stronger transformation value when they are embedded in administrative routines through actual technology use.

This interpretation also strengthens the connection between the proposed model and Dynamic Capabilities Theory. From this theoretical perspective, organizations do not transform only because they possess resources. They transform when they are able to mobilize, integrate, and reconfigure those resources in response to changing demands (Teece, 2018; Teece et al., 1997). In the public administration context, Santos (2024) argues that dynamic capabilities are relevant for understanding DT because public administration institutions must adapt routines, coordinate actors, and improve service delivery in changing environments. The present study extends this argument by identifying DCs, CM, and TA as complementary elements of that reconfiguration process. DCs provide the human resource base, CM provides the organizational support structure, and TA provides the operational channel through which resources are converted into digital transformation outcomes.

The study also contributes to digital governance theory by framing DT as an institutionally mediated process rather than a purely technological outcome. Prior studies have shown that DT in public administration is shaped by institutional inertia, bureaucratic culture, social construction processes, and organizational readiness (Norling, 2025; Tinjan, 2025). The present study supports this view and adds that DT depends on the alignment of people, processes, technologies, and institutional support. Therefore, Digital Governance Capacity helps explain why the availability of digital technologies does not automatically produce transformation. Technologies need to be accepted, used, organized, and embedded within public administration routines.

Another theoretical contribution concerns the role of DIP. Previous studies have shown that digital inequality affects access to online government services, citizen engagement, and the public value of e-government (Singh & Chobotaru, 2022; Tran Pham & Le Hoang Thuy To Nguyen, 2024; Wong & Ho, 2022). In this study, DIP was included as the inclusion-oriented dimension of Digital Governance Capacity. Although the moderating effect of DIP was not statistically significant, this result provides a useful theoretical clarification. It suggests that DIP may operate more strongly in citizen-facing models than in internal organizational models focused on public servants. Thus, DIP remains conceptually relevant, but its effect may depend on the level of analysis. In internal public administration processes, DT appears to be more directly associated with CM and TA. In citizen-facing digital government models, DIP may become more relevant for explaining access, inclusion, engagement, and public value creation.

The Lima-based context also strengthens the theoretical contribution of the study. Public administration institutions in Lima operate under a common national digital transformation framework, but they differ in mandates, resources, technological infrastructure, and implementation capacity. This context allows the study to examine Digital Governance Capacity in an emerging economy where digital reform is institutionally promoted, but unevenly implemented. Therefore, the study contributes to the literature by adapting digital governance and dynamic capabilities perspectives to a Latin American public administration setting characterized by institutional heterogeneity, bureaucratic routines, and unequal levels of digital readiness.

Overall, the contribution of this study is not the proposal of isolated relationships among known variables. Its contribution is the theoretical integration of these variables into a Digital Governance Capacity framework. This framework explains DT as the result of an institutional alignment process in which DCs provide the human foundation, CM creates the organizational conditions, TA operates as the implementation mechanism, and DIP represents the inclusion-oriented dimension of digital governance. In this way, the study extends technology adoption theory, refines Dynamic Capabilities Theory in the public administration context, and contributes to digital governance literature by explaining how public administration institutions in emerging economies convert digital capabilities and change practices into digital transformation outcomes.

6.4. Policy and Managerial Implications

The policy and managerial implications of this study should be interpreted with caution because the research design is cross-sectional and non-experimental. Even so, the observed relationships offer useful guidance for public managers and policymakers. DT strategies should not focus only on technology acquisition. They should also include structured CM, continuous training, internal communication, and support mechanisms that help civil servants use digital tools in daily administrative routines.

From a managerial perspective, digital skills development should not be treated as an isolated training activity. Therefore, the results should not be interpreted as suggesting that training alone is sufficient to drive digital transformation. Rather, training becomes effective when it is embedded within broader organizational and governance capabilities, including change management, leadership support, technology adoption, and process redesign. The results suggest that DCs are more useful when they are connected to implementation plans, leadership support, and opportunities to use digital tools in real work processes. Public managers should link digital training with process redesign, system use, and service delivery improvement.

Public institutions should also prioritize the quality of TA. This includes selecting user-oriented platforms, promoting interoperability, supporting employees during implementation, and monitoring whether digital tools are actually incorporated into administrative routines. It is not enough to introduce technology symbolically. Institutions should promote actual use, continuity, and integration into daily work.

CM should be included as a formal component of digital reform strategies. Public institutions should develop communication plans, internal leadership support, participation mechanisms, implementation guidance, and resistance management strategies. These actions are especially important in bureaucratic environments, where rigid structures can slow down or weaken digital initiatives (Norling, 2025; Tinjan, 2025).

From a policy perspective, DT strategies should follow an integrated approach. Policies should simultaneously develop DCs, strengthen CM, and promote effective TA. Isolated interventions may be less useful for supporting meaningful digital modernization. This integrated approach is relevant in emerging public administrations, where institutional fragmentation and uneven digital readiness may limit the sustainability of digital reform (Espina-Romero, 2025).

Although DIP did not moderate the central relationship, it remains relevant for policy. The absence of a significant moderating effect does not mean that institutional digital inclusion practices are unimportant. Rather, it suggests that, in this internal public administration model, DIP may not directly alter how TA translates into DT. Even so, digital inclusion practices remain important for inclusive governance, equitable service delivery, and the long-term sustainability of public digitalization (Singh & Chobotaru, 2022; Tran Pham & Le Hoang Thuy To Nguyen, 2024).

6.5. International and Contextual Relevance

This study contributes empirical evidence from Lima, Peru, to international debates on DT in public administration. Although the findings should not be generalized to all Latin American public administrations, they provide evidence from an emerging economy context where public institutions face digital reform pressures, heterogeneous capacities, and bureaucratic constraints.

The results suggest that DT in emerging contexts is associated not only with digital skills, but also with the institutional capacity to organize those skills through CM and TA. This broadens the comparative discussion between developed and developing contexts. It also highlights the importance of analyzing DT as a process shaped by organizational readiness, public administration capabilities, and institutional constraints.

The Lima context is relevant because public institutions operate under the same national digital transformation framework but differ in resources, mandates, and implementation capacity. In this setting, digital governance capacity becomes a useful concept for interpreting how public organizations align people, processes, technologies, and inclusion-oriented practices in institutional modernization.

The study also supports a more realistic view of DT in emerging public administrations. Institutional transformation does not appear to depend on a single factor. Rather, it is linked to the alignment of DCs, CM, TA, and contextual conditions. This interpretation connects the findings with broader debates on public administration, digital governance, and state modernization.

6.6. Limitations and Future Research

This study has several limitations. First, the methodological design is cross-sectional and non-experimental. Therefore, the results should be interpreted as structural associations rather than causal relationships. Future research could use longitudinal designs to examine how the relationships among DCs, CM, TA, DT, and DIP evolve over time.

Second, the study used a perception-based survey. This approach is appropriate for examining public servants’ views, but it does not capture objective indicators of digital transformation, institutional performance, or service quality. Future studies could combine survey data with administrative records, digital service indicators, interviews, or case studies to provide deeper analytical evidence.

Third, the research focused on public servants in Lima. This may limit the generalization of the findings to other institutional, territorial, or national contexts. Public organizations may differ in administrative culture, technological infrastructure, leadership practices, and digital readiness. Future studies could replicate the model in other regions of Peru or in other emerging economy contexts.

Fourth, the model focused on five constructs: DCs, CM, TA, DT, and DIP. These constructs are relevant for examining digital governance capacity, but other factors may also matter. Future research could include strategic leadership, organizational culture, institutional capabilities, public value creation, data governance, or citizen trust.

Finally, future studies could examine DIP in citizen-facing models. The non-significant moderating role of DIP in this study suggests that DIP may not operate strongly within internal organizational processes. However, prior literature indicates that DIP may be more relevant in relationships between e-government use, citizen engagement, and public value (Singh & Chobotaru, 2022; Tran Pham & Le Hoang Thuy To Nguyen, 2024). This opens a useful path for future comparative research.

Future research should also further test the Digital Governance Capacity framework using longitudinal, comparative, and multilevel designs. Longitudinal studies would help examine how DCs, CM, TA, DT, and DIP evolve over time and how institutional capabilities are developed during different stages of digital reform. Comparative studies across regions, countries, or types of public administration institutions would allow researchers to identify whether the proposed framework operates similarly in different institutional and socioeconomic contexts. Multilevel studies could also distinguish between individual-level factors, such as civil servants’ DCs, organizational-level factors, such as CM and TA, and broader institutional or citizen-facing factors, such as DIP. These approaches would strengthen the theoretical development of Digital Governance Capacity and provide a more precise understanding of how public institutions convert digital capabilities and change practices into sustainable digital transformation outcomes.

7. Conclusions

This study examined the structural relationships among DCs, CM, TA, DT, and DIP in the public administration of Lima. The results show that DCs and CM are positively associated with TA, while CM and TA show significant positive relationships with DT. TA also mediates the relationships between DCs and DT, and between CM and DT. No significant moderating relationship was found for DIP in the association between TA and DT. These findings should be interpreted as structural associations, not as causal effects, because the study used a cross-sectional and perception-based design.

A relevant finding is that DCs show a direct and statistically significant relationship with DT (β = 0.160, p = 0.014), confirming H3. However, the small effect size suggests that DCs have limited explanatory relevance when considered alone. This finding indicates that digital skills are important, but they are not sufficient by themselves to explain DT in public administration.

The study contributes to the literature by framing DT as an institutionally mediated process linked to digital governance capacity. The findings suggest that public administration transformation is better understood through the alignment of DCs, CM, and TA, rather than through the isolated presence of digital skills or technologies. This contribution is relevant because it shows how human capabilities, organizational support, and technology use can be interpreted as interconnected dimensions of institutional modernization.

The strongest relationship observed in the model was between TA and DT. This suggests that the effective implementation and institutional use of technologies may be a central mechanism associated with DT in public administration. Overall, the findings indicate that DT is an integrated process linked to human capabilities, organizational mechanisms, effective technology use, and institutional efforts to promote digital inclusion.

DIP remains important as an institutional inclusion-oriented condition, even though its moderating role was not statistically significant in this internal organizational model. This result suggests that DIP may not directly strengthen the relationship between TA and DT within public administration institutions. However, digital inclusion practices remain relevant for inclusive governance, equitable service delivery, and the sustainability of public digitalization.

From a practical and policy perspective, DT initiatives should be designed in an integrated manner. Public institutions should combine the development of DCs, the strengthening of CM, support for TA, and the promotion of digital inclusion practices such as training, equitable access strategies, awareness campaigns, and the adaptation of digital tools to users with different levels of technological readiness. This alignment may contribute to more sustainable and inclusive outcomes in digital governance.