Review Reports - Digital Technology Usage and Family Farms’ Uptake of Green Production Technologies—Evidence from Citrus Family Farms in Jiangxi Province

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The current research operationalizes digital technology as the mean index of "pre-production information acquisition, in-production IoT/AI, and post-production e-commerce", without distinguishing the differential impacts of different types of digital technologies. The digital technology index can be decomposed into three independent variables: "pre-production digital information (X1), in-production digital management (X2), and post-production digital marketing (X3)", and re-regressed to identify the differential effects of various technologies.
Although the paper clearly states the mediating logic of "digital technology, perceived benefits, and technology adoption", it does not provide a detailed description of the internal formation path of "perceived benefits". For instance, is the improvement of "economic benefit perception" due to digital technology's "reduction of production costs" or "increase in product revenue"? The current analysis does not clearly distinguish between these two key paths, resulting in a somewhat general explanation of the mechanism.
The heterogeneity analysis reveals that "digital technology only significantly promotes technology adoption in small-scale family farms", but does not provide an in-depth explanation of the underlying reasons. Is it because small-scale farms are more lacking in "technical information" or more constrained by "capital"? The current conclusion remains at the level of "phenomenon description" and has not been deeply integrated with theories such as resource constraint theory, which limits the theoretical extensibility of the conclusion.
Add a subsection titled "Research Limitations and Generalization Boundaries" in the discussion section, clearly pointing out the sample characteristics of "Jiangxi citrus farms", and comparing the differences in digital technology application scenarios between citrus and other crops to avoid over-generalization of the conclusion.
The paper proposes a three-dimensional strategy of "digital empowerment, socialized services, and skills training", but does not propose differentiated policies based on the heterogeneity results. For example, low-capital farms may need more "subsidies for low-cost digital tools", and small-scale farms may need more "practical digital skills training". The current suggestions are somewhat general and lack operability.

Author Response

Dear reviewer:

We appreciate your careful work on our submission of "Digital Technology Adoption and Family Farms’ Uptake of Green Production Technologies—Evidence from Family Farms in Jiangxi Province" (Manuscript ID:sustainability-3918173). We are also grateful to the reviewers for their constructive comments. We have revised the manuscript following the editor's and reviewers' comments, and carefully proofread the manuscript to minimize typographical, grammatical, and bibliographical errors.

Revised portions are marked in revision mode in the paper. The main corrections in the paper and respondse to the reviewer’s comments are as follows:

Reviewer 1：

Comment 1: The current research operationalizes digital technology as the mean index of "pre-production information acquisition, in-production IoT/AI, and post-production e-commerce", without distinguishing the differential impacts of different types of digital technologies. The digital technology index can be decomposed into three independent variables: "pre-production digital information (X1), in-production digital management (X2), and post-production digital marketing (X3)", and re-regressed to identify the differential effects of various technologies.

Response 1: Thank you very much for your valuable feedback. In the revised manuscript, we will make the following adjustment: the original composite "Digital Technology Index" will be disaggregated into three independent variables—" pre-production digital information (X1), in-production digital management (X2), and post-production digital marketing (X3)"—which will be subjected to separate regression analyses.

Comment 2: Although the paper clearly states the mediating logic of "digital technology, perceived benefits, and technology adoption", it does not provide a detailed description of the internal formation path of "perceived benefits". For instance, is the improvement of "economic benefit perception" due to digital technology's "reduction of production costs" or "increase in product revenue"? The current analysis does not clearly distinguish between these two key paths, resulting in a somewhat general explanation of the mechanism.

Response 2: Based on the theoretical framework diagram and the reviewer's comments, here is the revised version of section 2.2. The revision provides a more detailed description of the internal formation paths of perceived benefits, clearly distinguishing between the pathways of "cost reduction" and "revenue increase" for economic benefit perception, while also refining the explanations for social and environmental benefit perceptions. Firstly, from the cost reduction perspective, digital production technologies such as IoT sensors, drones, and big data analytics enable precision input management. This precision directly lowers the consumption of water, fertilizers, pesticides, and labor. By providing real-time monitoring and data on input usage, digital tools make these cost savings quantifiable, tangible, and predictable, thereby strengthening the perception of economic benefits derived from lower production expenditures. Secondly, from the revenue increase perspective, post-production digital technologies, including e-commerce platforms and market information systems, provide farmers with direct access to market data. This allows them to identify price premiums for green products, understand consumer demand trends, and secure more favorable sales channels. By clarifying the market value and price advantages of green agricultural products, digital technology enhances farmers' perception of potential revenue enhancement, making the economic benefits more salient. The interplay of these two paths—reducing costs and increasing income—collectively shapes a robust perception of economic benefit.

Comment 3: The heterogeneity analysis reveals that "digital technology only significantly promotes technology adoption in small-scale family farms", but does not provide an in-depth explanation of the underlying reasons. Is it because small-scale farms are more lacking in "technical information" or more constrained by "capital"? The current conclusion remains at the level of "phenomenon description" and has not been deeply integrated with theories such as resource constraint theory, which limits the theoretical extensibility of the conclusion.
Response 3: We sincerely thank the reviewer for this insightful comment, which has helped us significantly strengthen the theoretical depth of our discussion. We have revised the manuscript to address the core question.

In the revised Discussion section (Section 5.1), we have moved beyond mere description by explicitly integrating Resource Constraint Theory [36] to explain the pronounced effect among low-capital farms. We theorize that for these farms, capital scarcity acts as a "binding constraint." Digital technologies, particularly modular and low-cost applications, serve as a "leapfrogging" solution. They provide super-marginal returns by delivering substantial efficiency gains and cost savings without the need for large upfront investments in physical assets, thereby directly alleviating the capital barrier.

Furthermore, we directly address the reviewer's specific question by distinguishing the primary constraints for different farm types. For small-scale farms, we posit that the significant effect is likely attributable to information asymmetry reduction, rather than capital being the sole or primary constraint. These smaller operations often lack access to formal technical extension services. Digital platforms directly bridge this information gap by providing accessible, low-cost knowledge, which lowers the cognitive and technical barriers to adoption. This explanation extends the application of concepts related to Information Cascade Theory in a digital agriculture context.

Comment 4: Add a subsection titled "Research Limitations and Generalization Boundaries" in the discussion section, clearly pointing out the sample characteristics of "Jiangxi citrus farms", and comparing the differences in digital technology application scenarios between citrus and other crops to avoid over-generalization of the conclusion.

Response 4: Thank you for your insightful comment. We have carefully considered your suggestion and have substantially revised the manuscript by adding a dedicated subsection titled "5.2 Research Limitations and Generalization Boundaries" in the Discussion chapter, revised section explicitly highlights the "geographical and crop specificity of our sample (citrus farms in Jiangxi Province)" as a key boundary condition for generalizing the findings. We explicitly caution against over-generalization by elaborating that "Citrus cultivation, often occurring in hilly terrain, has specific technological needs and management practices."Furthermore, we directly address your suggestion to compare application scenarios by stating: "The digital tools and their perceived benefits might differ significantly in application scenarios for field crops or other perennial crops, which may have more standardized production processes or different market structures."This addition serves to clearly position our findings within the context of Jiangxi's citrus production systems, while thoughtfully acknowledging that the drivers and effectiveness of digital technology could vary in fundamentally different agricultural contexts, such as large-scale field cropping or other perennial fruit systems. We conclude this point by proposing a clear direction for future research: "Future research should validate these findings across different cropping systems and agro-climatic regions to enhance the external validity and generalizability of the conclusions."

Comment 5: The paper proposes a three-dimensional strategy of "digital empowerment, socialized services, and skills training", but does not propose differentiated policies based on the heterogeneity results. For example, low-capital farms may need more "subsidies for low-cost digital tools", and small-scale farms may need more "practical digital skills training". The current suggestions are somewhat general and lack operability.

Response 5: We are deeply grateful for this insightful and constructive comment. The reviewer rightly points out that policy recommendations should be precisely tailored to the specific needs of different farmer groups identified through heterogeneity analysis. We fully agree that moving from general strategies to actionable, differentiated policies significantly enhances the practical value of our research.To address this comment thoroughly, we have substantially revised and expanded the Policy Implications section within the Discussion chapter. Digital empowerment should be differentiated: promoting low-cost, modular digital tools and service models for low-capital farms, and providing integrated system solutions with financial support for large-scale farms. Socialized services should be expanded, particularly digital technology outsourcing services to enhance accessibility across all farm types. Skills training must be targeted: focusing on practical digital literacy for small-scale farms to overcome information barriers, and on system management and data analytics for larger operations.

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for inviting me to review the manuscript entitled "Digital Technology Adoption and Family Farms’ Uptake of Green Production Technologies-Evidence from Family Farms in Jiangxi Province". This study focuses on the impact of digital technology adoption on family farms’ uptake of green production technologies in Jiangxi Province’s citrus - producing regions. It adopts a scientific research framework, uses Logit models and multiple testing methods, and draws meaningful conclusions. The research is of great significance for promoting the green transformation of agriculture and the application of digital technology in agriculture. However, there are still some deficiencies in the research, which need to be revised and improved to enhance the depth, comprehensiveness, and persuasiveness of the study.

(1) The study mentions using the Theory of Planned Behavior and the expected utility framework to explain the mechanism of digital technology affecting farmers’ adoption of green production technologies, but the integration is not in-depth enough. This makes the theoretical explanation of the study relatively one-sided and unable to fully explain it. Hence, please integrate the theory more into the research, and add variables related to subjective norm and perceived behavioral control into the research model to explain the mechanism. Also, it is necessary to strengthen the connection between the expected utility framework and the research content. Further clarify how digital technology affects farmers’ expected utility of green production technologies and enrich the theoretical basis of the study. Some studies are very help and can be referred to, such as 10.15244/pjoes/187165.

(2) The study measures the digital technology usage index by taking the mean of three dummy variables. However, this method treats the three links equally, ignoring their differences. In addition, the study does not consider the frequency and depth of digital technology use, which may lead to inaccurate measurement of the digital technology usage level. Please adopt a weighted average method to measure the digital technology usage index and determine the weight of each link according to their importance. At the same time, add indicators to measure the frequency and depth of digital technology use. Besides, please design a multi-item scale to measure perceived economic, social, and environmental benefits, and use statistical methods such as Cronbach’s α coefficient to test the reliability and validity.

(3) The measurement of perceived economic, social, and environmental benefits only relies on a single question, which may not fully reflect the true perception of farmers. Hence, please conduct in-depth analysis of the reasons for the differences in the promotion effect of digital technology in different sample groups. And expand the scope of heterogeneity analysis, adding farmers’ personal characteristics as grouping variables. Besides, please analyze the differences in the promotion effect of digital technology on green production technology adoption among farmers of different ages and education levels and explore the reasons for these differences.

(4) Although the study conducts heterogeneity analysis from the perspectives of capital level, land fragmentation degree and operation scale, it lacks in-depth discussion on the reasons for the differences. In addition, the study does not consider the heterogeneity of farmers’ personal characteristics, such as age and education level.

(5) The study mentions that the data is cross-sectional, the research scope of green production technologies is narrow, and the sample is limited to Jiangxi Province’s citrus-producing regions. However, the discussion of these limitations is too superficial, and there is no in-depth analysis of the impact of these limitations. In terms of future research directions, the study only puts forward general suggestions. Please conduct in-depth analysis of the impact of research limitations on the research results and put forward specific and feasible future research plans.

Comments on the Quality of English Language

the quality of English language should be improved and polished by the MDPI company.

Author Response

Dear reviewer:

Revised portions are marked in revision mode in the paper. The main corrections in the paper and respondse to the reviewer’s comments are as follows:

Reviewer 2:

Comment 1: The study mentions using the Theory of Planned Behavior and the expected utility framework to explain the mechanism of digital technology affecting farmers’ adoption of green production technologies, but the integration is not in-depth enough. This makes the theoretical explanation of the study relatively one-sided and unable to fully explain it. Hence, please integrate the theory more into the research, and add variables related to subjective norm and perceived behavioral control into the research model to explain the mechanism. Also, it is necessary to strengthen the connection between the expected utility framework and the research content. Further clarify how digital technology affects farmers’ expected utility of green production technologies and enrich the theoretical basis of the study. Some studies are very help and can be referred to, such as 10.15244/pjoes/187165.
Response 1: Thank you for this insightful comment regarding the need to deepen the theoretical integration of the Theory of Planned Behavior (TPB) and the Expected Utility Framework in our study. We fully agree that a more robust theoretical explanation strengthens the paper. We have revised the manuscript to address your valuable suggestions, focusing on the following key aspects:

As suggested, we have more explicitly and deeply integrated the core constructs of the TPB—specifically adding and elaborating on the roles of subjective norms and perceived behavioral control—within the expected utility framework. The revised theoretical framework in Section 2 now clearly posits that digital technologies influence farmers' adoption decisions by systematically shaping their attitudes ,subjective norms, and perceived behavioral control. These three factors collectively determine the farmer's overall expected utility of adopting a green production technology. A rational farmer is theorized to adopt when the expected utility, calculated based on these TPB components, is positive. This integration provides a more holistic and multi-dimensional explanation of the decision-making mechanism.Furthermore, we have strengthened the connection between digital technology and the expected utility framework by specifying the pathways. For instance, we elaborate on how digital platforms provide information that shapes attitudes towards economic and ecological benefits, how they make social expectations more salient through online communities and success stories, and how they enhance perceived control by offering precise management tools and reducing uncertainty. This directly clarifies how digital technology affects the key inputs into the farmer's expected utility calculation.

We sincerely appreciate the suggested reference. While our current survey data allowed us to operationalize and test mediators related to farmers' perceptions of economic, social, and environmental benefits. we acknowledge that comprehensively measuring and incorporating distinct scales for all TPB variables, particularly subjective norms and perceived behavioral control in their entirety, was beyond the scope of our existing dataset. We have added a note in the limitations section of the discussion to explicitly state this and to highlight your valuable suggestion as a critical direction for future research.

Comment 2: The study measures the digital technology usage index by taking the mean of three dummy variables. However, this method treats the three links equally, ignoring their differences. In addition, the study does not consider the frequency and depth of digital technology use, which may lead to inaccurate measurement of the digital technology usage level. Please adopt a weighted average method to measure the digital technology usage index and determine the weight of each link according to their importance. At the same time, add indicators to measure the frequency and depth of digital technology use. Besides, please design a multi-item scale to measure perceived economic, social, and environmental benefits, and use statistical methods such as Cronbach’s α coefficient to test the reliability and validity.
Response 2: Thank you for this insightful and constructive feedback. We fully agree that employing a weighted average method, incorporating indicators for usage frequency and depth, and using multi-item scales to measure perceived benefits would provide a more precise and comprehensive depiction of digital technology usage levels and their impact mechanisms. These are undoubtedly valuable directions for improvement.

However, we would like to explain that due to objective constraints of the survey data collected during the research period, the questionnaire was already fixed, it is challenging for us to make such fundamental adjustments to the core measurement model in the current revision. Altering the measurement method directly could lead to inconsistencies with the already collected data structure.

Although we are unable to implement all your suggestions directly in the current empirical analysis, we take these comments very seriously and plan to address them as follows:

We will add a dedicated paragraph in the "Discussion" section of the manuscript to explicitly acknowledge the aforementioned limitations in measuring the digital technology usage index and perceived benefits. We will cite your valuable recommendations directly, stating that adopting weighted averages, incorporating frequency/depth metrics, and using multi-item scales are important directions for future research. We will candidly explain that the current approach of using an equally weighted average and single-item measures was a constraint of the initial research design and acknowledge this as a limitation of the present study.

In the future research, we will add items to the questionnaire to measure the intensity and integration level of digital technology use in different production segments separately. For the three dimensions of perceived benefits (economic, social, environmental), we will design multiple measurement items each to form multi-item scales, and commit to rigorously testing the reliability and validity of these scales in subsequent studies.

Comment 3: The measurement of perceived economic, social, and environmental benefits only relies on a single question, which may not fully reflect the true perception of farmers. Hence, please conduct in-depth analysis of the reasons for the differences in the promotion effect of digital technology in different sample groups. And expand the scope of heterogeneity analysis, adding farmers’ personal characteristics as grouping variables. Besides, please analyze the differences in the promotion effect of digital technology on green production technology adoption among farmers of different ages and education levels and explore the reasons for these differences.
Response 3: Thank you for your valuable feedback regarding our heterogeneity analysis. Building upon the original heterogeneity analysis, we will incorporate farmers' personal characteristics as grouping variables to conduct more detailed grouped regression analyses. This will allow us to examine whether the impact of digital technology varies across different farmer profiles. Furthermore, based on these findings, we will delve deeper into the underlying reasons for these observed differences by integrating theoretical perspectives with the analysis of individual characteristics.

Comment 4: Although the study conducts heterogeneity analysis from the perspectives of capital level, land fragmentation degree and operation scale, it lacks in-depth discussion on the reasons for the differences. In addition, the study does not consider the heterogeneity of farmers’ personal characteristics, such as age and education level.
Response 4: We thank the reviewer for this insightful comment. In the revised manuscript, we will address this by:

1.Delving deeper into the potential reasons behind the observed differences across capital levels, land fragmentation, and operation scale. This will be achieved by integrating existing theoretical frameworks and empirical evidence into the discussion section.

2.Incorporating a new dimension of heterogeneity based on farmers' personal characteristics, specifically age and education level. We will conduct corresponding subgroup analyses and discuss how these factors may influence the effectiveness of digital technology.

Comment 5: The study mentions that the data is cross-sectional, the research scope of green production technologies is narrow, and the sample is limited to Jiangxi Province’s citrus-producing regions. However, the discussion of these limitations is too superficial, and there is no in-depth analysis of the impact of these limitations. In terms of future research directions, the study only puts forward general suggestions. Please conduct in-depth analysis of the impact of research limitations on the research results and put forward specific and feasible future research plans.
Response 5: We thank the reviewer for this valuable feedback. We have added a dedicated "Research Limitations and Generalization Boundaries" subsection (Section 5.2) to the Discussion, providing an in-depth analysis of three key limitations: the cross-sectional data, narrow technological focus, and sample specificity. Specifically, we elaborate on how the cross-sectional design limits causal inference and observation of dynamic behavioral changes, discuss the constrained generalizability of findings from capital-intensive water-fertilizer integration to labor-intensive technologies, and highlight how the terrain-specific characteristics of Jiangxi citrus cultivation affect broader applicability. Corresponding concrete future research directions are proposed, including longitudinal panel tracking, comparative studies across technology types, and validation in diverse agro-ecological zones.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Author.

There are some concerns on your writing of the paper.

Title suggest : Digital technology usage and Family Farms’ Uptake of Green Production Technologies - Evidence from citrus Family Farms in Jiangxi Province

Green Production Technologies is not properly introduce in first paragraph of the introduction. Better to explain the concept of green production technology.

In the introduction there is many statements without the citations. better to add the citations. (e.g. Line 137 -150 for theories)

Abstract : need to clearly mention your findings with numerically.

Line 151-154 - Delete the sentence

Line 213: H1-b: need to explain like H1-a

Statistical analysis part is missing in the methodology

Tables introduction is not completed (Table 4)

Table column headings are not clear to the readers (Table 2,3,4,5,6,)

Foot note of the tables are missing (Table 3,5)

Discussion part is need to improve with previous literature with the new findings. Need to improve the discussion.

Author Response

Dear reviewer:

Revised portions are marked in revision mode in the paper. The main corrections in the paper and respondse to the reviewer’s comments are as follows:

Reviewer 3:

Dear Author.

There are some concerns on your writing of the paper.

Comment 1: Title suggest : Digital technology usage and Family Farms’ Uptake of Green Production Technologies - Evidence from citrus Family Farms in Jiangxi Province
Response 1: Thank you for your positive feedback on our title modification. As suggested, we have revised the manuscript title.

Comment 2: Green Production Technologies is not properly introduce in first paragraph of the introduction. Better to explain the concept of green production technology.

Response 2: We sincerely thank the reviewer for this valuable suggestion. We have revised the introduction to provide a clear definition and classification of Green Production Technologies, as suggested. It is crucial to first define their target—green production technologies themselves, which refer to production methods and processes designed to reduce environmental pollution, lower resource consumption, and minimize ecological damage while maintaining or improving productivity. The revised text now explicitly defines Green Production Technologies and introduces the academic categorization into labor-intensive and capital-intensive technologies, citing relevant literature . This clarification provides a stronger conceptual foundation for the study and better frames our subsequent analysis. The modifications can be seen in the marked-up manuscript.

Comment 3: In the introduction there is many statements without the citations. better to add the citations. (e.g. Line 137 -150 for theories)
Response 3: Thank you for pointing out the lack of citations for some statements in the introduction. We fully agree with your comment and have accordingly added relevant literature citations to the introduction section to ensure academic rigor. For instance, we have included supporting references for the relevant theories to clarify the source of arguments and strengthen the justification.

He, P., Zhang, J. and Li, W. The role of agricultural green production technologies in improving low-carbon efficiency in China: Necessary but not effective.Journal of Environmental Management,, 2021, 293, p.112837.
Guo, Z. and Zhang, X. Carbon reduction effect of agricultural green production technology: A new evidence from China.Science of the Total Environment,, 2023, 874, p.162483.
Ilbery, B. W. Agricultural decision-making: a behavioural perspective.Progress in Human Geograph, 1978, 2(3), 448-466.
Shen Y, Shi R, Yao L, et al. Perceived value, government regulations, and farmers’ agricultural green production technology adoption: evidence from China’s Yellow River Basin. Environmental Management, 2024, 73(3): 509-531.
Li M, Wang J, Zhao P, et al. Factors affecting the willingness of agricultural green production from the perspective of farmers' perceptions. Science of the Total Environment, 2020, 738: 140289.
Xu D, Liu Y, Li Y, et al. Effect of farmland scale on agricultural green production technology adoption: Evidence from rice farmers in Jiangsu Province, China. Land Use Policy, 2024, 147: 107381.
Li J, Feng S, Luo T, et al. What drives the adoption of sustainable production technology? Evidence from the large scale farming sector in East China. Journal of Cleaner Production, 2020, 257: 120611.
Larcher M, Engelhart R, Vogel S. Agricultural professionalization of Austrian family farm households-the effects of vocational attitude, social capital and perception of farm situation. German Journal of Agricultural Economics (GJAE), 2019, 68(1): 28-44.

Comment 4: Abstract : need to clearly mention your findings with numerically.

Response 4: Thank you for your valuable feedback on our manuscript. We agree that explicitly stating the key numerical findings in the Abstract significantly enhances its clarity and impact. We have revised the Abstract accordingly.

Comment 5: Line 151-154 - Delete the sentence

Response 5: We have deleted the sentence at lines 151-154 as suggested. The original sentence was "Rigorous testing using Sobel-Bootstrap methods systematically reveals how digital technologies influence green technology decisions by enhancing multidimensional benefit expectations. Third, extending heterogeneity analysis.". Its removal has made the context more concise and coherent.

Comment 6: Line 213: H1-b: need to explain like H1-a

Response 6: We have expanded the explanation for hypothesis H1-b at line 213 to match the level of detail and clarity provided for H1-a. Specifically, Digital technology enables farmers to apply green production technologies on a larger scale by optimizing resource allocation, improving management efficiency, and reducing transaction costs. Green production technologies often require reaching a certain threshold in application area to fully realize their economic and ecological benefits. Digital technologies help farmers manage this complexity through precision equipment and big data analytics, optimizing inputs and operations for larger areas. Additionally, digital marketing platforms secure market outlets and reduce transaction costs, mitigating market risks and increasing the expected utilityof large-scale adoption. Therefore, this study proposes the following hypothesis: H1-b: Digital technology exerts a significant positive effect on the scale of farmers' adoption of green production technologies.

Comment 7: Statistical analysis part is missing in the methodology

Response 7: We thank the reviewer for pointing out this omission. In the revised manuscript, we will add a new subsection titled " Descriptive statistics " within the "Results" section.

Comment 8: Tables introduction is not completed (Table 4)

Response 8: This section details and analyzes the statistical results presented in Table 4.

Comment 9: Table column headings are not clear to the readers (Table 2,3,4,5,6,)

Response 9: We agree that clear column headings are essential for readers to understand the table content. We have revised the column headings in all relevant tables.

Comment 10: Foot note of the tables are missing (Table 3,5)

Response 10: We appreciate you pointing out this oversight. Necessary footnotes have been added to Tables 3 and 5.

Comment 11: Discussion part is need to improve with previous literature with the new findings. Need to improve the discussion.
Response 11: We sincerely thank the reviewer for this valuable suggestion. We fully agree that strengthening the dialogue between our new findings and the existing literature is crucial for highlighting the theoretical contribution of our study. We have thoroughly revised the Discussion section (Section 5) to address this comment.

Reviewer 4 Report

Comments and Suggestions for Authors

Congrats to the authors for a good and extensive work.

The article examines the impact of digital technology usage on farmers' adoption of water-fertilizer integration technology within green production practices. Basically, the authors use digital technology on a large scale providing a practical way to get over obstacles including technological complexity, budgetary constraints, and information asymmetry during the promotion.

The Introduction section is comprehensive on the subject and includes topics like: numerous researchers that analyzed the impact of digital technologies on green production technologies, providing a theoretical foundation for current research, China’s policies in promoting and improving digital technologies, family farms situation in China, green production technologies, direct and indirect effects of digital technology on green production technology adoption. There is also explained the personal contribution of the current study to the field.

The Research Methods section is based on survey data from 432 family farms in Jiangxi Province's primary citrus-producing regions, continuing the work conducted in 2023 by the Citrus Industry System Research Team at Jiangxi Agricultural University among citrus family farms in Jiangxi Province.

Based on regression results, the effect of digital technology use on farmers’ water-fertilizer integration adoption is presented, the impact of digital technology adoption on water-fertilizer integration and the impact of digital technology adoption on the timeline and scale of water-fertilizer integration implementation are analyzed. There were also performed robustness checks, endogeneity test, mechanism analysis and heterogeneity analysis to validate the results.

The results reveal that digital technologies significantly increase farmers' probability of adopting water-fertilizer integration technology, the duration of adoption, and the proportion of adopted area. This conclusion aligns with existing research on digital technologies promoting agricultural green development. More, this study provides evidence specific to family farms as operational entities, contributing fresh perspectives to research on digital technologies empowering agricultural green development.

There were also addressed a few limitations of the study.

The conclusions are clearly stated and supported by the findings.

The study includes an average number of specialized references (31), these being extremely current (100% from the last 5 years) and well chosen from research journals with a high impact factor.

Author Response

Dear reviewer:

Revised portions are marked in revision mode in the paper. The main corrections in the paper and response to the reviewer’s comments are as follows:

Reviewer 4：

Comment:

Congrats to the authors for a good and extensive work.

There were also addressed a few limitations of the study.

The conclusions are clearly stated and supported by the findings.

The study includes an average number of specialized references (31), these being extremely current (100% from the last 5 years) and well chosen from research journals with a high impact factor.

Response : We are profoundly grateful for your exceptionally positive and insightful review of our manuscript. Thank you for your generous congratulations and for recognizing our work as "good and extensive." We are deeply encouraged that you highlighted the comprehensive nature of our introduction, the robustness of our methodological approach based on survey data from 432 family farms, and the validity of our findings strengthened by various tests. Your comment that our study "provides evidence specific to family farms" and contributes "fresh perspectives" is particularly rewarding and affirms the core contribution we hoped to make. We also sincerely appreciate your positive note on the currency and quality of our references. Your thoughtful and encouraging feedback has been a great source of motivation for our team.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors conduct some revision and have some improvement. However, there are still some deficiencies in the research, which need to be revised and improved to enhance the depth, comprehensiveness, and persuasiveness of the study.

(1) In this study, there is a lack of the connection between the expected utility framework and the research content. Hence, further clarify how digital technology affects farmers’ expected utility of green production technologies and enrich the theoretical basis of the study; this study is helpful 10.15244/pjoes/187165.

(2) In this study, there is a lack of a multi-item scale to measure perceived economic, social, and environmental benefits. Hence, please use a scale and statistical methods such as Cronbach’s α coefficient to test the reliability and validity.

(3) Although the study conducts heterogeneity analysis from the perspectives of capital level, land fragmentation degree, and operation scale, it lacks in-depth discussion on the reasons for the differences. Hence, a better discussion should be added and enriched.

(4) The study mentions that the data is cross-sectional and the scope of green production technologies is narrow. However, the discussion of these limitations is too superficial, and there is no in-depth analysis. Please conduct an in-depth analysis of the impact of research limitations and put forward specific and feasible future research plans.

Comments on the Quality of English Language

The quality of the English language should be improved and polished by the MDPI company.

A certificate of polishing is necessary for next submission

Author Response

Comment 1: In this study, there is a lack of the connection between the expected utility framework and the research content. Hence, further clarify how digital technology affects farmers’ expected utility of green production technologies and enrich the theoretical basis of the study; this study is helpful 10.15244/pjoes/187165.

Response 1: We sincerely thank the reviewer for this insightful and constructive comment. We fully agree that strengthening the connection to the expected utility framework would significantly enhance the theoretical depth of our study. As suggested, we have carefully studied the recommended literature (Li et al., 2025) and revised our theoretical section (Section 2, "Theoretical Framework and Research Hypotheses") to explicitly integrate the expected utility framework. This reference has been immensely helpful in addressing the gap identified in our original manuscript.

Digital-technology-enabled perception of benefits essentially assists farmers in more clearly quantifying the potential outputs of adoption behavior, thereby forming more defined revenue expectations. This process of quantification and clarification is a critical step in translating subjective perceptions into calculable decision variables. By providing concrete data on cost savings, potential yield increases, and market price premiums, digital tools reduce the ambiguity surrounding the adoption of Green Production Technologies (GPTs).

The research by Li et al. (2025) corroborates that the information integration capacity of digitalization contributes to the formation of stable income expectations among farmers. Their findings demonstrate how digital platforms synthesize information on market trends, input costs, and climatic conditions, enabling a more reliable forecast of the financial returns from technological adoption. This stabilization of expected income directly enhances the perceived utility of adoption.

Comment 2: In this study, there is a lack of a multi-item scale to measure perceived economic, social, and environmental benefits. Hence, please use a scale and statistical methods such as Cronbach’s α coefficient to test the reliability and validity.

Response 2: We sincerely thank the reviewer for this valuable comment regarding the measurement of perceived benefits. The reviewer rightly emphasizes the importance of using multi-item scales and statistical rigor, such as Cronbach's alpha, to establish the reliability and validity of psychological constructs.

In our study, the measurement of perceived economic, social, and environmental benefits was indeed based on single-item indicators for each construct. This approach was primarily adopted due to constraints in questionnaire design and length during the large-scale survey, aiming to reduce respondent burden and ensure a higher response rate. Consequently, as the reviewer correctly pointed out, it is not methodologically possible to calculate internal consistency metrics like Cronbach's alpha for these single-item measures.

We fully acknowledge this as a limitation in our measurement approach. We have proactively addressed this issue in the revised manuscript within the "5.2 Research Limitations and Generalization Boundaries" section. We explicitly state that the use of single-item measures, while practical, may not fully capture the multidimensional nature of the perceived benefit constructs and limits the thorough assessment of their psychometric properties. We have added a specific plan in the limitations section, stating that subsequent studies should employ validated multi-item scales to measure these perceptual variables. This will allow for a more robust evaluation of reliability and validity and provide a finer-grained understanding of the psychological mechanisms.

Comment 3: Although the study conducts heterogeneity analysis from the perspectives of capital level, land fragmentation degree, and operation scale, it lacks in-depth discussion on the reasons for the differences. Hence, a better discussion should be added and enriched.

Response 3: We sincerely thank the reviewer for this invaluable suggestion. We completely agree that an in-depth theoretical interpretation of the heterogeneity results is crucial for enhancing the depth and academic value of our study. Following your advice, we have made additions to the "5.1 Theoretical Implications" section. We moved beyond merely restating the results and attempted to connect the observed phenomena to broader theoretical frameworks.

For capital levels, the stronger promotional effect observed among low-capital farms aligns with Resource Constraint Theory. This may be due to the fact that high-capital farms are more inclined to invest in mechanized or automated production system systems, where the marginal benefits of additional digital technology upgrades may be diminishing. Instead, for low-capital farms still relying on labor-intensive practices, digital tools enable them to bypass intermediate capital-intensive technologies and adopt more advanced and efficient solutions directly.

For the scale of operations, the significant impact of digital technology on small-scale farms highlights the role of digitalization in overcoming the ineconomies of scale in access to information. Large-scale farms usually have stronger capabilities and channels to obtain technical services and market information, making their absolute information disadvantage less serious. For small-scale farms, digital platforms significantly reduce the unit cost per acre of information, effectively compensating for their scale-related weaknesses and creating a more level playing field in terms of information access.

For land fragmentation, the positive impact of digital technology is more pronounced in groups with lower levels of land fragmentation. Contiguous land provides ideal conditions for the deployment of fixed or mobile digital infrastructure such as IoT sensors, automated irrigation networks, drone cruises, etc. Digital technology can be seamlessly integrated with large-scale and standardized operating processes, significantly improving agricultural production efficiency and saving labor and time costs.

Comment 4: The study mentions that the data is cross-sectional and the scope of green production technologies is narrow. However, the discussion of these limitations is too superficial, and there is no in-depth analysis. Please conduct an in-depth analysis of the impact of research limitations and put forward specific and feasible future research plans.

Response 4: We sincerely thank the reviewer for this profound and constructive comment. We fully agree that an in-depth analysis of the research limitations is crucial for demonstrating academic rigor and guiding future research. As suggested, we have thoroughly revised and enriched the "5.2 Research Limitations and Generalization Boundaries" section.

We have transformed vague suggestions for future research into a set of actionable and research agendas. Regarding data, We explicitly recommended the use of multi-period panel data in future studies. Regarding theoretical testing, We proposed concrete methods for building an integrated analytical framework (e.g., Structural Equation Modeling, SEM), suggested employing multi-item scales for more precise measurement of psychological constructs, and outlined plans for systematically comparing the roles of different psychological pathways. Regarding generalizability, We put forward specific ideas for treating technology type as a moderating variable in comparative studies across technologies, and outlined clear pathways for replication studies and multi-regional comparative analysis across different agro-ecological zones and dominant cropping systems to test and extend the external validity of our findings.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for addressing the points raised out.

Author Response

Dear Authors,

Comment : Thank you for addressing the points raised out.

Response : We sincerely thank you for your time and for acknowledging our revisions. We are very pleased that our responses and modifications to the manuscript have addressed the points raised.We appreciate the opportunity to have improved our work through this review process.

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

Good job!