From Digital Governance to Public Value: The Mediating Role of Digital HRM in Public Service Organizations
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors The manuscript examines the relationship between Digital Governance (DG) and Public Value (PV) in a large public-sector organization, proposing Digital HRM (DHRM) as a mediating mechanism. Using a cross-sectional survey (N=278) and regression-based mediation analysis with bootstrapping, the study reports significant direct and indirect effects of DG on PV.The topic is timely and relevant to ongoing debates on digital transformation in public administration. The proposed mediation framework is conceptually meaningful and potentially contributes to understanding the internal organizational mechanisms that translate governance capacity into public value outcomes. In the text of the paper, where the research gap is discussed, the following is stated:
Empirical research still needs clearer internal mechanisms that connect governance capacity to public value in large, distributed public service organizations.
This statement is too general. The theoretical setting of the research should clearly indicate what previous studies have found, what are the limitations of previous research and how this work and the research results presented in it can build on the existing literature.
Public value in the text is broadly defined, and can seem conceptually diffuse. It must be theoretically justified.
Public value reflects societal value produced through service quality, efficiency, transparency, inclusion, and trust.
The current definition of public value appears conceptually expansive, incorporating elements traditionally associated with service performance (quality, efficiency) as well as legitimacy-related constructs (transparency, trust). While these dimensions are individually relevant, the manuscript would benefit from a clearer theoretical justification for aggregating them into a single reflective construct.
The authors may consider aligning the operationalization more explicitly with established public value frameworks (e.g., Moore, 1995; Meynhardt, 2009; Bryson et al., 2014) to strengthen construct clarity. Section 2.9 Measurement Quality Criteria presents a reliability assessment based solely on Cronbach's α coefficient. Although the values obtained are high (α > 0.90), research requires a broader assessment of the validity and reliability of the measurement model.
It is advisable to include additional indicators:
- • confirmatory factor analysis (CFA) to check the factor structure,
- • composite reliability (CR),
- • average variance extracted (AVE) to assess convergent validity,
- • discriminant validity test (HTMT ratio or Fornell–Larcker criterion).
Section 1.4 Control Variables and Alternative Explanations theoretically lists potential control variables, but they are not included in the empirical regression models. Given that the research and analysis are based on a cross-sectional design, the inclusion of control variables would further enhance analytical rigor and reduce the risk of omitted variables.
I recommend testing at least the following control factors:
- • job level (frontline, management level),
- • self-assessment of digital skills,
- • length of service.
Although the mediation model with three key constructs (DG, DHRM, and PV) is theoretically grounded and analytically coherent, its structure remains relatively linear and does not include potential moderating mechanisms. In contemporary research on the digital transformation of the public sector, contextual factors that can strengthen or weaken the main relationships in the model are often examined.
In the specific case, it would be justified to consider the role of at least one moderating effect (eg the level of digital skills of employees, hierarchical position, seniority or organizational unit), in order to determine whether the mediation relationship is stable in different subgroups or depends on the context of the implementation of digital practices. Incorporating moderation would provide a deeper understanding of the conditions under which digital governance most effectively contributes to public value.
The current model provides a basic causal-mediational framework, but extending it to moderated mediation or at least testing for interaction effects would further strengthen the theoretical and empirical relevance of the work. Include newer sources (within the last five years) from the field of digital governance and public value, in order to position the manuscript more clearly in the contemporary research context.
Describe the sampling procedure more precisely, including the method of distribution of the questionnaire and, if available, the response rate, for greater transparency of the methodology.
Reduce repetitions in the discussion and additionally summarize the parts that repeat the findings already presented, so that the argumentation is more focused and convincing.
According to the mentioned reviews, in order to improve the theoretical and research setting of the work, it is necessary to make the mentioned corrections.
Author Response
Response to Reviewer 1
Comment 1: Research Gap Discussion
Reviewer's Comment: "Empirical research still needs clearer internal mechanisms that connect governance capacity to public value in large, distributed public service organizations." This statement is too general. The theoretical setting of the research should clearly indicate what previous studies have found, what are the limitations of previous research and how this work and the research results presented in it can build on the existing literature.
Our Response: We thank the reviewer for this important observation. We have substantially revised the research gap discussion in Section 2.7 (Synthesis and Research Gap) to provide a more detailed and specific account of previous findings and limitations. The revised text now explicitly identifies four key limitations in existing research: (1) the scarcity of empirical tests linking governance to value outcomes despite the integrative framework by Vigoda-Gadot and Mizrahi [41]; (2) the absence of mediation analysis in prior e-HRM and public value studies; (3) the focus on Western developed-country contexts; and (4) the lack of systematic discriminant validity assessment between governance and HR constructs. Each limitation is now connected to specific prior studies, demonstrating how our research builds on and extends existing literature. [Changes highlighted in yellow in Section 2.7]
Comment 2: Public Value Definition
Reviewer's Comment: Public value in the text is broadly defined, and can seem conceptually diffuse. It must be theoretically justified. The current definition of public value appears conceptually expansive, incorporating elements traditionally associated with service performance (quality, efficiency) as well as legitimacy-related constructs (transparency, trust). The authors may consider aligning the operationalization more explicitly with established public value frameworks (e.g., Moore, 1995; Meynhardt, 2009; Bryson et al., 2014) to strengthen construct clarity.
Our Response: We appreciate this constructive feedback on conceptual clarity. We have substantially revised Section 2.2 (Public Value in the Digital Era) to provide a more rigorous theoretical justification for our public value operationalization. The revised text now explicitly integrates Moore's [28] strategic triangle framework, Meynhardt's [27] psychological-value approach, Jorgensen and Bozeman's [22] comprehensive inventory of public values, and Bryson et al.'s (2014) distinction between substantive outcomes and procedural values. We clarify that our operationalization captures PV as a reflective construct encompassing service quality, efficiency, transparency, inclusion, and trust—dimensions that collectively represent the societal benefits that public organizations are authorized and expected to produce. This alignment with established frameworks provides a stronger theoretical foundation for our measurement approach. [Changes highlighted in yellow in Section 2.2]
Comment 3: Measurement Quality Criteria
Reviewer's Comment: Section 2.9 Measurement Quality Criteria presents a reliability assessment based solely on Cronbach's α coefficient. It is advisable to include additional indicators: confirmatory factor analysis (CFA), composite reliability (CR), average variance extracted (AVE), and discriminant validity test (HTMT ratio or Fornell–Larcker criterion). Given the relatively high correlation between DG and DHRM (r = 0.818), it is particularly important to empirically confirm that these constructs are clearly separable.
Our Response: We thank the reviewer for this methodologically rigorous suggestion. We have comprehensively revised Section 4.9 (Measurement Quality Criteria) to include all recommended validity assessments. The revised section now describes: (1) Cronbach's alpha for internal consistency; (2) Confirmatory Factor Analysis (CFA) to verify the three-factor structure; (3) Composite Reliability (CR) with threshold of ≥0.70; (4) Average Variance Extracted (AVE) for convergent validity (≥0.50); and (5) HTMT ratio for discriminant validity (<0.85). Additionally, we have added a new validity results table (Table 2b) reporting CR, AVE, and HTMT values for all constructs, along with CFA model fit indices. These additions are particularly important given the high DG-DHRM correlation and provide empirical confirmation that our constructs are empirically distinct. [Changes highlighted in yellow in Section 4.9 and new Table 2b]
Comment 4: Control Variables
Reviewer's Comment: Section 1.4 Control Variables and Alternative Explanations theoretically lists potential control variables, but they are not included in the empirical regression models. I recommend testing at least: job level, self-assessment of digital skills, and length of service.
Our Response: We appreciate this suggestion to strengthen analytical rigor. We have revised the manuscript to include control variables in the empirical models. Specifically, we added Section 4.5 (Control Variables) in the Methods section describing the three controls: job level (frontline vs. managerial), self-assessed digital skills, and organizational tenure. We also added analysis results showing that these controls did not substantially alter the main relationships, thereby supporting the robustness of our mediation model. The indirect effect through DHRM remained significant (a×b=0.268; 95% CI [0.142, 0.412]) when controls were included. [Changes highlighted in yellow in new Section 4.5 and enhanced results section]
Comment 5: Sampling Procedure
Reviewer's Comment: Describe the sampling procedure more precisely, including the method of distribution of the questionnaire and, if available, the response rate, for greater transparency of the methodology.
Our Response: We have substantially revised Section 4.6 (Data Collection Procedures) to provide detailed information about the sampling procedure. The revised text now specifies: (1) the multi-channel distribution approach (email invitations, paper-based surveys via HR coordinators, and QR codes); (2) the data collection period (March-April 2025); (3) the number of questionnaires distributed (approximately 500); (4) the response rate (59.6% for returned surveys, 55.6% for usable responses); (5) the reminder protocol; and (6) non-response analysis results. These additions significantly enhance the transparency and replicability of our methodology. [Changes highlighted in yellow in Section 4.6]
Comment 6: Repetitions and Newer Sources
Reviewer's Comment: Reduce repetitions in the discussion. Include newer sources (within the last five years) from the field of digital governance and public value.
Our Response: We have revised the discussion section to reduce repetitive statements and have updated the literature review to include recent sources. The revised manuscript now incorporates several 2024-2025 publications including Vigoda-Gadot and Mizrahi [41] (2024), Hassan [19] (2025), Thu et al. [38] (2025), Okorie and De Guzman [32] (2025), and Mwangi and Kamau [30] (2024). We have streamlined the discussion to focus on novel contributions rather than reiterating findings already presented in the results section. [Changes highlighted in yellow throughout the manuscript]
Reviewer 2 Report
Comments and Suggestions for AuthorsThe conducted research is important because it demonstrates how digital governance can contribute to the creation of public value, but only when it is supported by appropriate digital human resource management (HRM) practices. The implementation of digital HRM practices that develop employees’ competencies, institutionalize digital routines, and support the stability of public service delivery makes it possible to translate governance assumptions into tangible benefits.
However, several weaknesses of the manuscript can be identified.
It is not entirely clear whether the authors treat the term e-HRM as a synonym of DHRM, or whether they differentiate between these concepts. This distinction does not emerge clearly from the text. The title of section 2.3, “Digital HRM and e-HRM: mechanisms for capability and service quality,” explicitly suggests that the authors understand these two terms differently.
The authors state: “This sample size is adequate for the study’s mediation testing using regression and bootstrapping, and it also supports reliable estimation of internal consistency and correlation-based validity diagnostics for multi-item constructs.” (lines 296-299). What are the substantive or methodological grounds for this claim?
If the questionnaire was prepared in Arabic (line 310), for what purpose was it reviewed by bilingual academics/practitioners? Did the respondents complete the questionnaire in a language other than Arabic? The text indicates that the questionnaire items were adapted from established scales in the literature (lines 310-311), but there is no information about the specific sources of these scales or which items were adopted or modified. The absence of such references makes it difficult to assess the theoretical foundations of the measurement instrument and the transparency of the variable operationalization procedure.
Furthermore, what exactly did the pilot study, referred to in the title of the section (line 309), involve? According to standard methodological assumptions, it should have been conducted among postal employees. Additionally, what did the “quality control” described in section 4.7 (lines 321-327) consist of?
The manuscript also repeats the same information several times, for example that 278 complete questionnaires were obtained (lines: 296, 323, 365, 371). Are such repetitions truly necessary?
Author Response
Response to Reviewer 2
Comment 1: e-HRM vs DHRM Distinction
Reviewer's Comment: It is not entirely clear whether the authors treat the term e-HRM as a synonym of DHRM, or whether they differentiate between these concepts. This distinction does not emerge clearly from the text. The title of section 2.3, "Digital HRM and e-HRM: mechanisms for capability and service quality," explicitly suggests that the authors understand these two terms differently.
Our Response: We thank the reviewer for highlighting this ambiguity. We have clarified our terminology in the revised manuscript. Section 2.3 has been retitled to "Digital HRM: mechanisms for capability and service quality" to remove the implication of two distinct concepts. We have added an explicit statement explaining that we use e-HRM and digital HRM (DHRM) interchangeably to refer to the application of digital systems to HR processes across the HR value chain. We acknowledge that some scholars distinguish between e-HRM (digitization of administrative processes) and digital HRM (encompassing advanced analytics and AI-enabled functions), but clarify that this distinction is not critical for our theoretical framework. [Changes highlighted in yellow in Section 2.3]
Comment 2: Sample Size Justification
Reviewer's Comment: What are the substantive or methodological grounds for this claim about sample size adequacy?
Our Response: We have substantially revised Section 4.3 to provide comprehensive methodological justification for the sample size. The revised text now specifies four grounds: (1) Hair et al.'s [18] recommendation of minimum 200 for SEM with three constructs; (2) Preacher and Hayes' [34] guidance that 100-200 samples are sufficient for regression-based mediation with bootstrapping; (3) the 10 observations per parameter threshold for CFA (24 items × 10 = 240 minimum); and (4) power analysis showing adequate statistical power (>0.80) to detect medium effect sizes. These references provide solid methodological support for our sample size. [Changes highlighted in yellow in Section 4.3]
Comment 3: Questionnaire Language and Sources
Reviewer's Comment: If the questionnaire was prepared in Arabic, for what purpose was it reviewed by bilingual academics/practitioners? The text indicates that the questionnaire items were adapted from established scales in the literature, but there is no information about the specific sources of these scales or which items were adopted or modified.
Our Response: We have revised Sections 4.4 and 4.5 to address these concerns. We now explain that the bilingual review was conducted as part of Brislin's [8] back-translation method to ensure semantic equivalence between the original English scales and the Arabic translation. Regarding scale sources, we have added detailed information specifying that: DG items were adapted from Vigoda-Gadot and Mizrahi [41]; DHRM items were adapted from Wang et al. [42] and El-Emary et al. [14]; and PV items were adapted from Meynhardt [27] and Jorgensen and Bozeman [22]. All scale items were reviewed for face validity by subject-matter experts prior to data collection. [Changes highlighted in yellow in Sections 4.4 and 4.5]
Comment 4: Pilot Study and Quality Control
Reviewer's Comment: What exactly did the pilot study involve? What did the "quality control" described in section 4.7 consist of?
Our Response: We have substantially revised both sections to provide complete transparency. Section 4.5 now details the pilot study: conducted with 30 Egypt Post employees (not included in the final sample), assessing clarity, reliability, and face validity. Participants provided feedback on item wording, response scale clarity, and survey length. Based on pilot feedback, two items were reworded, and completion time was confirmed at 12-15 minutes. Preliminary reliability was α > 0.85 for all constructs. Section 4.7 now describes the multi-step quality control process: exclusion of incomplete responses (>10% missing), identification and exclusion of straight-lining patterns (2 cases), review of response times (1 case excluded for <5 minutes), and handling of demographic missing values. [Changes highlighted in yellow in Sections 4.5 and 4.7]
Comment 5: Repetitions
Reviewer's Comment: The manuscript also repeats the same information several times, for example that 278 complete questionnaires were obtained (lines: 296, 323, 365, 371). Are such repetitions truly necessary?
Our Response: We have reviewed the manuscript and removed redundant repetitions of the sample size information. The number of valid cases (278) is now mentioned only in Section 4.3 (sample description) and Section 5.1 (sample profile), eliminating unnecessary repetition in other sections. This streamlining improves the manuscript's conciseness and readability. [Changes highlighted in yellow throughout]
Reviewer 3 Report
Comments and Suggestions for AuthorsThis manuscript examines the relationship between digital governance (DG) and public value (PV), with digital human resource management (DHRM) proposed as a mediating mechanism in public service organizations. Based on survey data collected from employees of Egypt Post and analyzed through regression-based mediation with bootstrapping, the study finds that DHRM partially mediates the effect of DG on PV. The topic is timely and relevant in the context of public-sector digital transformation, and the empirical analysis appears methodologically sound. The findings offer potentially meaningful implications for understanding how digital governance initiatives translate into organizational value creation. However, the manuscript requires substantial revisions in terms of theoretical positioning, structural coherence, and clarity of contribution.
(1)The abstract is not fully structured and lacks conceptual completeness. While it presents methodological procedures and statistical findings in detail, it does not sufficiently articulate the broader research background, the specific research gap, the theoretical positioning of the study, or its explicit contributions. A well-developed abstract should clearly state the motivation, theoretical framework, research problem, methodology, key findings, and implications in a balanced manner. The authors should substantially revise the abstract to ensure it reflects the full intellectual structure of the study rather than primarily reporting results.
(2)The introduction does not clearly specify the marginal contribution of the study. Although it discusses the importance of digital governance, public value, and digital HRM, it does not convincingly demonstrate what gap in the existing literature this research addresses or how it advances theory. The manuscript would benefit from a more explicit articulation of its theoretical contribution, whether in terms of mechanism identification, integration of research streams, contextual extension, or methodological refinement. Without such clarification, the study risks appearing incremental rather than theory-enhancing.
(3)The hypotheses form a coherent mediation framework, yet no conceptual model diagram is provided. Given that the study centers on a structured relationship among DG, DHRM, and PV, a visual representation of the theoretical model is necessary. Including a conceptual framework figure that illustrates the hypothesized paths and mediating mechanism would significantly improve clarity, enhance logical transparency, and align the manuscript with established conventions in empirical management and public administration research.
(4)The structural organization of the manuscript requires improvement. The section titled “Control variables and alternative explanations” should be integrated into the Methods section rather than presented separately. Control variables and model specification are methodological components and should be discussed as part of research design. Reorganizing this content will improve structural coherence and ensure logical consistency between theoretical development, research design, and empirical testing.
(5)The managerial implications are currently presented in a general and somewhat abstract manner. The recommendations should be more directly derived from the empirical findings and reflect the specific mechanisms identified in the study. Since the results indicate a partial mediation effect of DHRM, the implications should clarify how public organizations can strategically translate digital governance into workforce capabilities and routines through targeted HR digitalization initiatives. Strengthening the alignment between findings and managerial recommendations would enhance the practical relevance and scholarly rigor of the manuscript.
Author Response
Response to Reviewer 3
Comment 1: Abstract Structure
Reviewer's Comment: The abstract is not fully structured and lacks conceptual completeness. While it presents methodological procedures and statistical findings in detail, it does not sufficiently articulate the broader research background, the specific research gap, the theoretical positioning of the study, or its explicit contributions.
Our Response: We have completely restructured the abstract to provide conceptual completeness. The revised abstract now clearly articulates: (1) Background - the digital transformation context; (2) Research Gap - the lack of integrated governance-HRM-value studies; (3) Theoretical Framework - socio-technical systems theory and resource-based view; (4) Purpose - the mediation mechanism; (5) Methodology - including the new validity assessments; (6) Findings - statistical results; (7) Contributions - three distinct theoretical contributions; and (8) Implications - practical recommendations. This structure ensures a balanced presentation of the full intellectual structure of the study. [Changes highlighted in yellow in the Abstract]
Comment 2: Marginal Contribution
Reviewer's Comment: The introduction does not clearly specify the marginal contribution of the study. It does not convincingly demonstrate what gap in the existing literature this research addresses or how it advances theory.
Our Response: We have added a new paragraph at the end of the Introduction section that explicitly states the three distinct contributions of our study: (1) integrating the previously separate research streams of digital governance and digital HRM by specifying DHRM as a mediating mechanism; (2) extending public value scholarship to an under-researched Middle Eastern public-sector context (Egypt Post); and (3) offering a validated measurement framework for future research on the governance-HRM-value nexus in public organizations. This addition clearly articulates how our research advances theory beyond incremental contributions. [Changes highlighted in yellow in Section 1]
Comment 3: Conceptual Model Diagram
Reviewer's Comment: The hypotheses form a coherent mediation framework, yet no conceptual model diagram is provided. A visual representation of the theoretical model is necessary.
Our Response: We note that the original manuscript did include Figure 1 (Research model - mediation framework) showing the hypothesized relationships between DG, DHRM, and PV. This figure has been retained in the revised manuscript and shows the direct and indirect paths (H1, H2, H3, H4) in the mediation framework. The figure aligns with standard conventions in empirical management and public administration research for presenting mediation models.
Comment 4: Structural Organization
Reviewer's Comment: The section titled "Control variables and alternative explanations" should be integrated into the Methods section rather than presented separately. Control variables and model specification are methodological components.
Our Response: We have reorganized the manuscript structure as suggested. The control variables discussion has been moved from Section 3.4 to a new Section 4.5 within the Methods section. This reorganization improves structural coherence and ensures logical consistency between theoretical development, research design, and empirical testing. The control variables are now presented alongside other methodological components such as measures, sampling, and analysis procedures. [Changes highlighted in yellow in new Section 4.5]
Comment 5: Managerial Implications
Reviewer's Comment: The managerial implications are currently presented in a general and somewhat abstract manner. The recommendations should be more directly derived from the empirical findings and reflect the specific mechanisms identified in the study.
Our Response: We have substantially revised the managerial implications section to provide more specific, actionable recommendations directly tied to our findings. The revised implications now include: (1) specific strategies for aligning digital learning pathways with citizen-facing service standards; (2) guidance on using HR analytics to identify capability gaps; (3) recommendations for implementing digital performance management; (4) suggestions for expanding employee self-service while maintaining fairness perceptions; and (5) guidance on establishing ethical guardrails for HR analytics including audit trails and bias monitoring. Each recommendation is explicitly linked to our partial mediation finding and the dual-path logic (governance → value through both direct and HR-mediated pathways). [Changes highlighted in yellow in the Discussion section]
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have specifically and comprehensively addressed the comments raised in my review. I find these clarifications and the corrections they have made satisfactory.
Reviewer 3 Report
Comments and Suggestions for AuthorsI recommend accepting it in its present form.
