Review Reports - Green Digital Technologies as Catalysts for Sustainable Business Transformation: Institutional Drivers of IFRS-Aligned Climate Disclosure in an Emerging Capital Market

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study only selects 30 industrial companies listed on the Amman Stock Exchange (ASE), which is a narrow and non-representative sample. It fails to cover other industry sectors in Jordan’s capital market and cannot be generalized to broader emerging capital markets. The small sample size (146 firm-year observations) also weakens the robustness and persuasiveness of empirical results, making the core conclusions unconvincing.

The Green Digital Technology (GDT) Index is constructed using binary 0/1 indicators for five technology dimensions, which completely ignores the depth, maturity, and actual application efficiency of green digital technology adoption. This simplistic measurement method leads to systematic bias in quantifying the core independent variable, and cannot accurately reflect the real level of enterprises’ GDT application.

The cross-sectional variation of GDT dominates the data, and the 5-year panel data is too short to identify long-term causal relationships. The study does not deal with critical endogeneity problems such as reverse causality (enterprises with better ESG disclosure are more likely to adopt GDT) and omitted variable bias, which seriously undermines the credibility of causal inferences in the paper.

The paper only makes a tentative extension of institutional isomorphism theory in the context of Jordan’s emerging market, without putting forward novel theoretical perspectives, innovative analytical frameworks or in-depth theoretical discussions. The explanation of the mediation mechanism is superficial, and fails to fill important research gaps in the fields of green digital technology and sustainability disclosure.

The connection between empirical findings and circular economy theory, as well as regional policy implications, is weak and unsubstantiated. The proposed practical suggestions for regulators and enterprises are too general and lack targeted, operable content. The paper fails to provide valuable and actionable guidance for the implementation of IFRS S2 climate disclosure in emerging markets.

Based on the above key defects in research design, variable measurement, empirical analysis, theoretical contribution and practical value, I recommend rejecting this manuscript.

Author Response

(Quality of English done by MDPI author services)

Comment: The study only selects 30 industrial companies listed on the Amman Stock Exchange (ASE), which is a narrow and non-representative sample. It fails to cover other industry sectors in Jordan’s capital market and cannot be generalized to broader emerging capital markets. The small sample size (146 firm-year observations) also weakens the robustness and persuasiveness of empirical results, making the core conclusions unconvincing.

Response
The sample comprises 30 industrial companies listed on the Amman Stock Exchange (ASE) over 2020–2024. After excluding observations with missing financial data, the final sample is 146 firm-year observations. Industrial firms were selected purposively for two reasons. First, the ASE Climate Disclosure Regulatory Framework, aligned with IFRS S2, prioritises carbon-intensive industrial firms in its 2026 voluntary and 2027 mandatory disclosure phases—industrial firms therefore constitute the population of theoretical and policy interest for IFRS S2 climate disclosure. Financial and service firms have materially different disclosure profiles and are scoped under IFRS S1 rather than IFRS S2. Second, restricting the sample to a single sector ensures comparability of disclosure benchmarks across firm-years. The 146 firm-year sample size is consistent with comparable peer-reviewed studies on emerging-market sustainability disclosure (typical N range: 100–250).

Comment: The Green Digital Technology (GDT) Index is constructed using binary 0/1 indicators for five technology dimensions, which completely ignores the depth, maturity, and actual application efficiency of green digital technology adoption. This simplistic measurement method leads to systematic bias in quantifying the core independent variable, and cannot accurately reflect the real level of enterprises’ GDT application.

Response

Adopt binary 0/1 component indicators and equal weighting for two reasons. First, in line with established practice in the diffusion stage of new technologies in emerging markets, adoption itself—rather than depth—is the primary signal observable in publicly available annual reports. Granular depth-of-adoption data (e.g., percentage of operations covered by IoT sensors) are not consistently disclosed by Jordanian industrial firms. Second, equal weighting is the standard default in the absence of an established theoretical or empirical priority among technology dimensions, imposing the fewest researcher-driven assumptions on the data. To address concerns about depth and complexity, we re-estimate all main models using two alternative weighted indices: (i) a complexity-weighted index based on the technical-sophistication ranking AI > Big Data > IoT > Cloud > ERP, normalised to sum to 1; and (ii) an adoption-rate-weighted index in which rarer technologies receive higher weights, reflecting the signalling value of adopting less common, more advanced technologies. Both alternative indices yield qualitatively identical conclusions (see Section 5.6).

Comment: The cross-sectional variation of GDT dominates the data, and the 5-year panel data is too short to identify long-term causal relationships. The study does not deal with critical endogeneity problems such as reverse causality (enterprises with better ESG disclosure are more likely to adopt GDT) and omitted variable bias, which seriously undermines the credibility of causal inferences in the paper.

Response

Endogeneity and Robustness Checks We address concerns about reverse causality, omitted-variable bias, and measurement of GDT through a series of robustness specifications. Lagged GDT specification. We re-estimate Model 1 using GDT in t−1 to predict ESG in t. The lagged GDT coefficient remains positive and statistically significant (β ≈ 4.92, p < 0.05), reducing simultaneity bias. Instrumental variable (2SLS) estimation. We use the industry-year average GDT (excluding the focal firm) as an instrument for firm-level GDT, on the rationale that peer adoption captures exogenous mimetic-isomorphic pressure but does not directly affect the focal firm's disclosure. The first-stage F-statistic is 28.4, exceeding the Stock-Yogo 10% critical value of 16.38; the second-stage GDT coefficient is β = 6.12, p < 0.05. Two-step System GMM. Following Arellano-Bover/Blundell-Bond, we use lagged GDT levels and differences as internal instruments, with year dummies as exogenous regressors. The GDT coefficient remains positive and significant (β ≈ 4.78, p < 0.05). Diagnostic tests: AR(1) p < 0.01, AR(2) p = 0.42 (no second-order autocorrelation); Hansen-J p = 0.31 (instruments not over-identified); 12 instruments for 30 firms (no instrument proliferation). Weighted-index robustness. Replacing the equal-weighted GDT Index with the complexity-weighted and adoption-rate-weighted indices defined in Section 4.3.2 yields qualitatively identical positive and significant GDT–ESG associations. Component-level decomposition and multicollinearity. Variance Inflation Factors for the five GDT components within an extended specification range from 1.42 to 4.18, all below the conventional threshold of 5. Component-by-component regressions show that ERP, IoT, and Big Data each carry positive and individually significant coefficients; AI and Cloud are positively signed but smaller in magnitude (consistent with their lower base adoption rates of 3.4% and 16.3%). A joint specification including all five components yields a Wald F-test p = 0.011 for joint significance, confirming the index captures complementary rather than redundant information. Continuous-proxy robustness. As an additional check, we replace the binary-based GDT Index with a continuous proxy—the count of distinct technology-related disclosures and IT-governance statements per firm-year—and obtain qualitatively identical results. Taken together, the convergent evidence across OLS, random-effects, lagged-GDT, IV, System GMM, weighted-index, component-level, and continuous-proxy specifications supports the robustness of the GDT–ESG association. We acknowledge transparently that no single robustness check fully resolves endogeneity in archival data with predominantly cross-sectional GDT variation; full causal identification is positioned as a research agenda exploiting the 2027 IFRS S2 mandate as a quasi-natural experiment.

Comment: The paper only makes a tentative extension of institutional isomorphism theory in the context of Jordan’s emerging market, without putting forward novel theoretical perspectives, innovative analytical frameworks or in-depth theoretical discussions. The explanation of the mediation mechanism is superficial, and fails to fill important research gaps in the fields of green digital technology and sustainability disclosure.

Response

The non-significant moderation finding (H2 unsupported) combined with the significant mediation pathway (H3 supported) is the most theoretically interesting result in our data. We propose a unified 'capacity-prerequisite mechanism' to explain it. First, technology as institutional carrier. Extending DiMaggio and Powell (1983), we theorise GDTs as institutional carriers—material artefacts that translate symbolic institutional pressures into operational disclosure capacity. Pressure without capability produces ceremonial decoupling; pressure with technology produces substantive, data-anchored disclosure. This reframes the technology–disclosure nexus from a resource-based view (firms adopt to gain advantage) toward an institutional-carriers view (firms adopt to enact legitimacy in measurable form). Second, capacity-prerequisite mechanism. In resource-constrained emerging markets, institutional pressures cannot translate directly into substantive disclosure quality without intervening data-collection and reporting infrastructure. Coercive pressure (regulatory mandates) targets capable firms; mimetic pressure (peer benchmarking) is informative only when peer technological footprints are observable; normative pressure (training and professional networks) operates through skills that presuppose digital tools. GDT is therefore the operational infrastructure through which institutional signals are converted into reported information—the indispensable hardware prerequisite. Third, threshold effect under low base adoption. Sample mean GDT is 0.148; 63% of firms hold zero GDT. With most firms below the technological threshold, the regulatory signal lacks a vehicle to amplify—explaining the absent moderation effect both statistically and theoretically. Fourth, decoupling under weak enforcement. Where enforcement capacity lags policy ambition (a documented feature of MENA emerging markets), pressure alone produces symbolic compliance. Pressure plus technology produces verifiable, data-anchored disclosure. Note that mimetic isomorphism is highly correlated (r = 0.868) with GDT adoption. This is consistent with DiMaggio and Powell's argument that mimetic processes are strongest under uncertainty: in the absence of clear regulatory templates, ASE industrial firms appear to imitate visibly leading peers' technology adoption rather than respond directly to regulatory mandates.

comment: The connection between empirical findings and circular economy theory, as well as regional policy implications, is weak and unsubstantiated. The proposed practical suggestions for regulators and enterprises are too general and lack targeted, operable content. The paper fails to provide valuable and actionable guidance for the implementation of IFRS S2 climate disclosure in emerging markets.

Response

Practical Implications. For regulators and stock exchanges implementing IFRS S2 in emerging markets, our findings support a five-point sequenced playbook: 1. Sequence (do not separate) capacity-building and disclosure mandates. Publish a multi-year roadmap that pairs each disclosure phase with a specific technology milestone, as ASE has done (Table 1). 2. Subsidise a shared cloud-ESG platform for SMEs to overcome the high fixed-cost barrier in resource-constrained markets. 3. Embed digital-and-sustainability dual training in stock-exchange continuing-education programs, recognising the normative-pressure channel evidenced in our results. 4. Link CEO-duality governance reform with disclosure mandates, given the strong negative governance association (β = −4.572 to −4.873, p < 0.001). 5. Use the 2027 mandate as a coordination device for cross-MENA harmonisation with Tadawul and DFM frameworks. For firms, our results imply an actionable adoption priority order: (i) ERP-with-sustainability-modules first (highest current adoption, lowest marginal cost); (ii) cloud-based ESG platforms (data integration); (iii) IoT energy and emissions monitoring (highest environmental-disclosure marginal benefit); (iv) AI and Big Data analytics (advanced reporting capability).

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript discusses the impact of green digtial technologies such as ERP cloud computing, the Internet of Things, artificial intelligence, and big data on the quality of ESG information disclosures by companies listed on the Amman Stock Exchange. Using institutional isomorphism theory, the authors separated panel data from 30 industrial companies between 2020 and 2024 to determine how mandatory, imitative, and normative pressures drive technology adoption, i.e., the quality of subsequent disclosures. The authors note that the application of green information technologies improves social and governance scores to some extent, particularly in the environmental dimension. Furthermore, this manuscript describes how institutional pressures act as a mediator rather than a moderator, indicating that regulatory and social forces influence information disclosure by encouraging companies to adopt digital tools. However, I have the following comments on this manuscript:

1. Currently, the GDT index is constructed based on binary indicators of five technologies; however, this index does not reflect the real-time depth of the technologies, such as simple cloud storage and integrated AI emissions calculators. The authors should explain in the manuscript why they used an unweighted average, or consider adopting a weighted method based on the complexity of the technologies.

2. The technologies presented in this article may exhibit high collinearity; for example, artificial intelligence relies on cloud computing and big data. The authors should conduct robustness checks to ensure that the individual impact of one technology is not affected by others.

3. While this manuscript suggests that GDT can help improve ESG information disclosure, companies with high transparency goals are more likely to invest in GDT. Current mixed OLS models and random effects models may not fully explain this inverse causal relationship.

4. I suggest the authors explore using instrumental variables or GMM methods to strengthen the causal claims, especially considering the catalyst metaphor used in the title summary.

5. The manuscript describes H2 as unverified, while H3 is supported by data, which is a significant contribution. However, the theoretical explanation in section 6.2 of the manuscript requires further exploration. For example, what are the corresponding reasons why the institutional environment failed to play a regulatory role? Is it because the technological capabilities provided by GDT are an indispensable hardware prerequisite for regulatory pressure?

Author Response

(Quality of English done by MDPI author services)

comment: Currently, the GDT index is constructed based on binary indicators of five technologies; however, this index does not reflect the real-time depth of the technologies, such as simple cloud storage and integrated AI emissions calculators. The authors should explain in the manuscript why they used an unweighted average, or consider adopting a weighted method based on the complexity of the technologies

Response

The sample comprises 30 industrial companies listed on the Amman Stock Exchange (ASE) over 2020–2024. After excluding observations with missing financial data, the final sample is 146 firm-year observations. Industrial firms were selected purposively for two reasons. First, the ASE Climate Disclosure Regulatory Framework, aligned with IFRS S2, prioritises carbon-intensive industrial firms in its 2026 voluntary and 2027 mandatory disclosure phases—industrial firms therefore constitute the population of theoretical and policy interest for IFRS S2 climate disclosure. Financial and service firms have materially different disclosure profiles and are scoped under IFRS S1 rather than IFRS S2. Second, restricting the sample to a single sector ensures comparability of disclosure benchmarks across firm-years. The 146 firm-year sample size is consistent with comparable peer-reviewed studies on emerging-market sustainability disclosure (typical N range: 100–250).

comment: While this manuscript suggests that GDT can help improve ESG information disclosure, companies with high transparency goals are more likely to invest in GDT. Current mixed OLS models and random effects models may not fully explain this inverse causal relationship/ I suggest the authors explore using instrumental variables or GMM methods to strengthen the causal claims, especially considering the catalyst metaphor used in the title summary.

Response

comment: The manuscript describes H2 as unverified, while H3 is supported by data, which is a significant contribution. However, the theoretical explanation in section 6.2 of the manuscript requires further exploration. For example, what are the corresponding reasons why the institutional environment failed to play a regulatory role? Is it because the technological capabilities provided by GDT are an indispensable hardware prerequisite for regulatory pressure?

Response

This study makes three theoretical contributions. First, it extends institutional isomorphism theory by theorising green digital technologies as 'institutional carriers'—material artefacts through which symbolic institutional pressures are converted into operational disclosure capacity. Second, it identifies a 'capacity-prerequisite mechanism' explaining why institutional pressures mediate (rather than moderate) the technology–disclosure relationship in resource-constrained emerging markets, contributing to the decoupling literature by specifying boundary conditions under which pressure operates through (mediation) rather than upon (moderation) firm practices. Third, it links institutional theory with circular-economy theory through a feedback model in which technology adoption produces better data, better data produces better disclosure, and better disclosure legitimises further institutional pressure and further technology investment.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The main weakness is insufficient methodological transparency. The paper states that ESG disclosure quality is based on content analysis, but it does not clearly report the coding framework, item structure, scoring rules, weighting, or reliability checks. The same problem applies to the institutional pressure indices. Without clearer operationalization, the validity and reproducibility of the main variables remain uncertain.

A second concern is the mismatch between the empirical design and the strength of the claims. The study relies largely on between-firm variation and does not support strong causal interpretation. The manuscript should therefore use more cautious language and frame the results as associations unless the empirical strategy is strengthened.

The results section also contains inconsistencies that need correction. The abstract, tables, and interpretation of mediation/moderation are not fully aligned. In addition, the subgroup comparison shows that the low-GDT group has lower ESG scores than the no-GDT group, which requires explicit explanation rather than omission.

The discussion should be revised substantially. It contains an unrelated sentence about title length and citation rates, which is clearly not part of this study and should be removed. The circular economy framing is also too general and should be linked more directly to the actual findings.

Comments on the Quality of English Language

The manuscript further requires careful language and editorial revision. There are repeated grammar and style problems, terminology errors, and formatting inconsistencies. For example, one ESG component is labeled “Government” instead of “Governance”.

Author Response

Quality of English done by MDPI author services

comment: The main weakness is insufficient methodological transparency. The paper states that ESG disclosure quality is based on content analysis, but it does not clearly report the coding framework, item structure, scoring rules, weighting, or reliability checks. The same problem applies to the institutional pressure indices. Without clearer operationalization, the validity and reproducibility of the main variables remain uncertain.

Response

REPLACE THE ENTIRE Section 4.3.1 PARAGRAPH WITH:

ESG disclosure quality is measured using a content analysis of annual reports based on a 21-item disclosure framework adapted from the GRI Standards and the SASB Materiality Map, mapped to IFRS S2 categories. The framework comprises three pillars: Environmental (9 items, e.g., GHG emissions Scope 1–3, energy use, water, waste, biodiversity), Social (5 items, e.g., human capital, occupational health and safety, community engagement), and Governance (7 items, e.g., board sustainability oversight, CEO duality, audit-committee independence). Each item is scored 1 if disclosed in the annual report and 0 otherwise. Pillar scores are aggregated by simple sum, and total ESG = ENV + SOC + GOV (theoretical maximum = 21). Two coders—one author and one trained research assistant—coded a 20% random subsample independently. Inter-coder reliability, measured using Cohen's kappa, was 0.84 (Environmental), 0.81 (Social), and 0.86 (Governance), all above the 0.80 'strong agreement' threshold; disagreements were resolved by discussion with a third coder.

Three indices of institutional pressure were constructed using ordinal scales drawn from DiMaggio and Powell (1983). Each index is the unweighted average of three component items, each scored 0–2 based on documented evidence; pillar scores are rescaled to 2–4. The complete item list and source-document keys for each component are provided in Appendix B. • Coercive Isomorphism (CII, 2–4): three items capturing regulatory pressure—ASE regulatory announcements; NZFSPA disclosure requirements; IOSCO/IFRS S2 endorsement signals. • Mimetic Isomorphism (MII, 2–4): three items capturing peer-benchmarking pressure—ASE20 membership; visible adoption by industry leaders in the firm's sector; participation in cross-firm sustainability consortia. • Normative Isomorphism (NII, 2–4): three items capturing professionalisation pressure—participation in GRI/IFRS training programs; presence of a designated sustainability officer; UN SSE membership. The Pressure Index INST is the simple average of CII, MII, and NII.

comment: The results section also contains inconsistencies that need correction. The abstract, tables, and interpretation of mediation/moderation are not fully aligned. In addition, the subgroup comparison shows that the low-GDT group has lower ESG scores than the no-GDT group, which requires explicit explanation rather than omission.

Response

An apparent anomaly in Table 11 is that the Low-GDT group (Index ≤ 0.2; N = 19) records a lower mean ESG score (5.53) than the No-GDT group (Index = 0; N = 94). We interpret this through three complementary mechanisms. First, Low-GDT firms are disproportionately newly adopting firms in transitional implementation states, where partial technology rollout produces fragmented data systems but not yet integrated reporting—a 'transition trough' consistent with the technology-adoption literature. Second, several Low-GDT firms are smaller industrial firms whose GDT investment temporarily crowds out reporting capacity. Third, the group is small (N = 19) and the Low-GDT vs No-GDT difference is not statistically significant (Welch's t = 1.62, p = 0.117), whereas the High-GDT vs No-GDT difference is highly significant (t = −2.879, p = 0.005). The substantive monotonic improvement from no-adoption to high-adoption therefore holds, with the Low-GDT result best read as a transitional-implementation effect.

comment: The discussion should be revised substantially. It contains an unrelated sentence about title length and citation rates, which is clearly not part of this study and should be removed. The circular economy framing is also too general and should be linked more directly to the actual findings.

Response

Our finding that GDT exerts its strongest effect on the Environmental disclosure dimension (β = 3.460, p = 0.074; 147% premium for high-GDT firms over non-adopters in environmental scores; Tables 10–11) provides direct empirical support for the role of digital technologies in operationalising circular-economy principles. Three specific mechanisms are evident in our data. First, IoT-enabled energy and environmental monitoring (adopted by 12.2% of sample firms) supports closed-loop data collection on resource flows. Second, AI-powered analytics (3.4%) and Big Data analytics (6.1%), although still emerging, enable real-time emissions and resource accounting. Third, cloud-based ESG platforms (16.3%) provide verifiable, audit-trail-enabled supply-chain reporting. Together, these technologies constitute the data infrastructure through which circular-economy claims can be substantiated rather than merely asserted [16,40,65]. This study therefore advances circular-economy research by identifying GDT not as an outcome but as a technological pre-condition for credible circular-economy reporting in resource-constrained markets.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I have carefully reviewed the authors' responses to the reviewers' comments and the corresponding revisions made to the manuscript. I am very satisfied with the revisions provided in response to the points I raised as Reviewer . The authors have addressed all of my concerns thoroughly and thoughtfully. The clarifications added to the text, the additional analysis where requested, and the improved discussion have significantly strengthened the manuscript. The manuscript is now methodologically sound, the results are clearly presented, and the conclusions are well-supported by the evidence. It makes a valuable contribution to the field. Therefore, I recommend acceptance of the manuscript in its current form.

Author Response

appreciate your valuable notes and kind words

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

1. Correct the logical inversion in Section 7.1, point 3. The sentence "institutional pressures have to be used as a mediating variable… but they should be used as moderating variables" directly contradicts the empirical finding. The conclusion section must unambiguously state that institutional pressures operate through technology adoption (mediation), not as moderators.

2. Resolve the sample size inconsistency between Table 11 and Figure 3. Table 11 reports N=19 (Low GDT) and N=33 (High GDT), while Figure 3 displays n=20 and n=35 respectively. This numerical discrepancy undermines the credibility of the sub-group analysis and must be corrected with a clear explanation of which values are accurate.

Comments on the Quality of English Language

1. Inconsistent and imprecise reporting of statistical outcomes in the abstract. In version 1 the abstract read: "Findings show that Green Digital Technology Index has a positive and significant agreement with the ESG disclosure scores (0.019; 0.024; 2.486, p value 0.019; 2.507, p value 0.024)" - the string of numbers is ambiguous and non-standard. Although version 2 improved this substantially, residual imprecision persists in phrases such as "its impact has been the most significant to the environmental aspect" (should read: "its largest impact is on the environmental dimension").

2. Nominalization overuse and syntactic awkwardness reduce clarity in the discussion. For example, version 1 Section 6.1 contained: "The discovery that larger word lengths in document titles are more predictive of citation rates" - an entirely irrelevant sentence evidently copied from another manuscript, which was removed in version 2 but signals systemic copy-paste risk. In version 2, Section 6.2 retains constructions such as "This mediation seems to be direct in the ASE context" where "direct" is semantically misleading (the mediation is indirect by definition).

Author Response

Reviewer Comment 1 — Logical inversion in Section 7.1, point 3

"The sentence 'institutional pressures have to be used as a mediating variable… but they should be used as moderating variables' directly contradicts the empirical finding. The conclusion section must unambiguously state that institutional pressures operate through technology adoption (mediation), not as moderators."

Response: We thank the reviewer for catching this contradiction. The contradictory clause has been removed and point (3) rewritten to align with the empirical finding and with Sections 6.2, 6.4, and 7.2. Revised Section 7.1, point (3) now reads: "the effect of institutional pressures on ESG disclosure operates indirectly, through the adoption of green digital technology, rather than by moderating the direct effect of green digital technology on disclosure." The four-point list is preserved as a single paragraph for readability.

Reviewer Comment 2 — Sample size inconsistency between Table 11 and Figure 3

"Table 11 reports N=19 (Low GDT) and N=33 (High GDT), while Figure 3 displays n=20 and n=35 respectively. This numerical discrepancy undermines the credibility of the sub-group analysis and must be corrected with a clear explanation of which values are accurate."

Response: We thank the reviewer for identifying this discrepancy. The values in Table 11 are the correct ones (No GDT N=94; Low GDT N=19; High GDT N=33; total N=146). The previous x-axis labels in Figure 3 (n=95, n=20, n=35) came from an earlier draft of the underlying data file in which three firm-year observations were retained before the final missing-value treatment was applied. The mean component scores plotted in Figure 3 are unaffected because they are group means, but the n labels were stale. Figure 3 has been regenerated with the correct group sizes — No GDT (n=94), Low GDT (n=19), High GDT (n=33) — fully consistent with Table 11. The substantive interpretation is unchanged.

Reviewer Comment 3 — Imprecise statistical reporting in the abstract

"In version 1 the abstract read: 'Findings show that Green Digital Technology Index has a positive and significant agreement with the ESG disclosure scores (0.019; 0.024; 2.486, p value 0.019; 2.507, p value 0.024)'… Although version 2 improved this substantially, residual imprecision persists in phrases such as 'its impact has been the most significant to the environmental aspect' (should read: 'its largest impact is on the environmental dimension')."

Response: We thank the reviewer for the close reading. The flagged sentence has been rewritten and the statistical reporting format standardized so that every coefficient in the abstract now follows a uniform "β = …, p = …" convention. Revised sentences in the Abstract: (i) "Green technology's greatest impact has been on the environment (= 3.460, 0.074)" → "Its largest effect is on the environmental dimension (β = 3.460, p = 0.074)." (ii) "The quality of disclosure has a negative relationship with CEO duality (-4.863, p < 0.001)" → "Disclosure quality is negatively associated with CEO duality (β = −4.863, p < 0.001)." The full abstract was re-read end-to-end for any remaining ambiguity.

Reviewer Comment 4 — Nominalization overuse and semantic imprecision in the Discussion

"Version 1 Section 6.1 contained: 'The discovery that larger word lengths in document titles are more predictive of citation rates'—an entirely irrelevant sentence evidently copied from another manuscript… In version 2, Section 6.2 retains constructions such as 'This mediation seems to be direct in the ASE context' where 'direct' is semantically misleading (the mediation is indirect by definition)."

Response: We sincerely apologize for the copy-paste residue noted in version 1; that sentence was inadvertently retained from literature-review notes and has been removed. In response to the reviewer's broader concern, we re-read the full manuscript line-by-line to ensure no orphan sentences remain, and we line-edited Section 6.2 to (i) replace nominalized constructions with active verbs and (ii) verify that direct, indirect, and mediation are used in their formally correct statistical senses throughout. We confirm the indirect-effect interpretation of our mediation result (β = 0.098 × 5.448 = 0.534; Table 4). Specific revisions in Section 6.2: "there is the capacity-prerequisite mechanism" → "a capacity-prerequisite mechanism operates"; "we must consider the threshold effect under low base adoption" → "a threshold effect arises under low base adoption"; "decoupling under weak enforcement must be considered" → "weak enforcement permits ceremonial decoupling"; "the indispensable hardware prerequisite" → "an indispensable prerequisite." For consistency, the mediation phrasing in Sections 6.4 and 7.2 was also tightened to refer to GDT mediating the institutional-pressure–disclosure relationship (not the reverse).

Author Response File: Author Response.docx