Impact of University Social Responsibility on Strengthening Sustainable Rural Women’s Entrepreneurship: Multigroup Analysis Based on the SEM

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The reviewers raised the following questions, and the authors are invited to carefully consider.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

The English quality of the manuscript is relatively low, with many errors and non-standard wording.

Author Response

1. Introduction: The conclusion of the literature survey should be clearly stated by the author, explaining the role and significance of your research.

Despite growing evidence that multidimensional USR interventions—that is, those that simultaneously provide financial education, business training, and social support networks—produce the most consistent benefits for rural women's businesses, the existing literature is fragmented. Much of it is derived from qualitative case studies or single-purpose microfinance programs, and rarely models rich causal pathways or mediating effects of empowerment with robust quantitative approaches. Additional longitudinal analysis suggests that the durability of the findings depends on long-term institutional support; however, it is rarely modeled how such support relates to business survival and value chain integration.

This paper addresses these gaps by applying multi-group structural equation modeling to a 6-month post-experimental study of USR in rural Peru. Our equation setting simultaneously captures the direct effects of USR and the indirect effects, channeled through economic empowerment, on sustainable business entrepreneurship. In this way, it (i) integrates economic, social, and psychological empowerment into a single analytical framework, (ii) quantifies the relative contribution of each of the three components of USR to ultimate lasting sustainability, and (iii) provides an evidence-based roadmap for universities, policymakers, and development practitioners pursuing gender-responsive rural innovation ecosystems. It therefore advances theory on USR and rural entrepreneurship while informing widely scalable, multidimensional practical interventions.

2. Pages 4-6, Section- 2. Materials and Methods: Can the author summarize the survey results into a table to illustrate, so that readers can have a clearer understanding of the survey process and conclusions.

Table 1. Survey implementation and key pre-/post‑test results.

Dimension / Indicator	Specification	Outcome (Pre‑test)	Outcome (Post‑test)	Change
Survey phase	Baseline (Month 0) vs. Follow‑up (Month 6)	–	–	Two‑wave panel
Respondents (n)	Rural women, Chepén & Pacasmayo, aged 18–60	60 completed (100 %)	Same cohort	–
Instrument	33 Likert‑type items; 7 latent constructs; face‑to‑face	α = 0.78–0.92	α = 0.80–0.93	Reliability upheld
Analysis method	SmartPLS 4; 5 000‑sample bootstrap; multigroup SEM	–	–	Consistent
ICT Quality → ICT Impact	Standardised path β (p)	0.911 (0.000)	0.863 (0.000)	Stable, slightly
Community Participation → Economic Opportunities	β (p)	0.775 (0.000)	0.735 (0.000)	Stable
Accessibility → ICT Quality	β (p)	0.025 (0.795)	0.347 (0.001)	+0.322
ICT Content → ICT Quality	β (p)	0.505 (0.001)	0.209 (0.244)	−0.296
Self‑Esteem → ICT Quality	β (p)	0.309 (ns)	0.076 (ns)	−0.233
Significant pre‑post difference	Bootstrap MGA Δβ (p)	–	–	Accessibility → ICT Quality: Δβ = 0.322, p = 0.033

Note. Positive change indicates that the trajectory gained strength and statistical significance after the intervention.

 

3. Page 7, Section- 3. Results: Figure 1 needs to be modified to reduce its size, and the calibration of the data can be relocated. Check for similar situations throughout the text and make revisions.

View manuscript.

4. Page 10, Section- 3. Results: The author re annotated the format of Table 3, which is currently unsightly.

Path (pre → post)	β Difference (Post–Pre)	p (1-tailed)	p (2-tailed)
Accessibility → ICT Quality	0.322	0.016	0.033
Autonomy → ICT Quality	0.199	0.215	0.429
Community Participation → Economic Opportunities	−0.040	0.618	0.763
ICT Content → ICT Quality	−0.296	0.905	0.189
ICT Impact → Community Participation	0.051	0.327	0.655
ICT Quality → ICT Impact	−0.048	0.746	0.508
Self-Esteem → ICT Quality	−0.233	0.852	0.295

 

5. Pages 7-12, Section- 3. Results；Please ask the author to explain the three different data narrative processes more clearly, using different representations such as comparative figures. The current expression method repeats the original format multiple times, and the reviewer spends a lot of time reading and understanding it.

3.1. Pre-implementation Model of University Social Responsibility Programs

The pre-implementation structuring of USR revealed an initial ecosystem in which technology and social factors exhibit a series of interrelationships. The most significant relationships include: -ICT Quality → ICT Impact (β=0.863, p<0.001). As explained previously, any technological intervention is effective only to the extent that the technical quality of the solution is sufficient. -ICT Impact → Community Participation (β=0.741, p<0.001), communities that perceive benefits from technology are more willing to participate in the process. Community Participation → Access to Economic Opportunities (β=0.735, p<0.001), community participation is a social fabric through which communities access economic opportunities. ICT Accessibility → ICT Quality (0.347, p=0.001), accessibility significantly influences the perception of the term quality. Self-esteem and ICT content showed weaker or non-significant relationships, implying areas for future improvement actions.

As observed in Table 1, the statistical analysis supports the significant relationships identified in the pre-intervention model. It is particularly important that the relationships between ICT quality and impact (t= 23.751, p < 0.001), ICT impact and community partic-ipation (t= 8.168, p= 0.050 and t= 11.617 p< 0.001), and participation and access to eco-nomic opportunities (t = 11.617, p < 0.001), with standardized coefficients above 0.690. On the other hand, the p-values for technology accessibility; t= 0.259, p = 0.795, autonomy; t=0.841, p = 0.400 and self-esteem t=1.597, p= 0.111, are all at a non-significant level, con-firming the need for additional interventions in these fields. Therefore, the first round of the initial configuration allowed the identification of the hierarchical pattern of commu-nity development in which the quality of technology leads to an impact that influences participation and finally generates economic opportunities. The strong trend in the rela-tionship between quality and impact indicates that the communities were not simple re-cipients of technologies; instead, they were sensitive to the quality of the tools provided to them.

3.2. Post-implementation Model of USR Programs

After the implementation of the USR program, substantial changes in structural relationships are described as follows: ICT Quality → ICT Impact (β=0.911, p<0.001), this relationship, in addition to being maintained, was strengthened due to maturation in the use and consumption of ICT. Community Participation → Access to Economic Opportunities (β=0.775, p<0.001), the social mechanisms that enabled it were strengthened. ICT Content → Implemented ICT Quality (β=0.505, p=0.001); however, this relationship increased in terms of significance, as it was originally relevant, implying that the community matured technologically. Inverse relationship of ICT Impact with Community Participation (β=0.690, p<0.001) remained relevant but decreased, suggesting that the community had established participation mechanisms that do not depend as much on technology. The self-esteem coefficient showed an increase (β=0.309) despite not reaching statistical significance, suggesting the near importance of psychological factors.

For its part, Table 2 presents robust statistical evidence of the transformation in the post-intervention model. In particular, the t-statistical values of the relationships between ICT quality and impact, impact and community participation, and participation and accessibility to economic opportunities remain high. However, several t-values for these and other relationships are different. It is significantly new that access to ICT is a determinant of ICT quality t = 3.276, p = 0.001, and ICT quality is now a determinant of the dependent variable, autonomy. t = 2.034, p = 0.042. Therefore, the current post-intervention model suggests that the causal mechanism behind USR actions has changed significantly in the study community. In other words, the community is now more sophisticated and sees the strengthening of some relationships along with the emergence of other significant new ones. This may mean that, from the beginning, USR achieved a profound change in the way the community uses the technological and social resources at its disposal.

3.3. Comparative Analysis: Pre vs. Post-implementation

The Bootstrap MGA multigroup analysis yielded a single statistically significant difference between the two periods: the relationship between ICT accessibility and ICT quality (difference = 0.322, p = 0.033). This finding is revealing, as it shows that the USR program has overcome the barrier of technological access and, instead, has allowed the community to focus on more complex aspects of business and social development. The maintenance and improvement of community bonding and empowerment to economy from β = 0.863 to β = 0.911 and between β = 0.735 to β = 0.775 indicate that the USR program has preserved and improved the functional mechanisms prior to its implementation. The most relevant changes are: ICT content emerges as a judgment factor instead of a measurement one between β = 0.209 to β = 0.505. ICT Accessibility loses importance. Regarding Self-esteem, its role becomes more relevant from β = 0.076 to β = 0.309. These changes are indicative of a greater community orientation towards digital maturity, that is, from concentration on more contingent technological access to quality and content, as well as a more holistic orientation that incorporates psychosocial factors.

 

6. Pages 13-15, Section- 4. Discussion: The data analysis in the discussion section is unclear, and there is no real explanation of the reasons for this situation, nor does it reflect the innovation and scientific of the research. It is just a simple explanation of the data. The expression method of the manuscript is too single, which makes it difficult for reviewers to understand how this monotonous way reflects the progressiveness and research value of SCI articles.

View manuscript.

 

7. Pages 14-15, Section- 5. Conclusions: The expression method of the conclusion is incorrect. The conclusion section is a deeper study and analysis of the innovative and key content of the research, rather than a repetition of the original content. The manuscript also did not mention the shortcomings of this research work and the content for future research.

View manuscript.

 

8. Pages 16-17, Section- References: The format of references is incorrect, with some errors and insufficient quantity For example: Ahmed, M.; Yousaf, H.Q.; Naseer, M.; Rehman, S. The Role of Social Entrepreneurship Education and Corporate Social Responsibility in Shaping Sustainable Behaviour in the Education Sector of Lahore, Pakistan. Industry and Higher Education 2024, doi:10.1177/09504222241297538.

View manuscript.

9. Pages 1-17, Section- paper: The English language used in the manuscript is inaccurate, 
and there are errors in the sentence structure and commonly used writing formats. The entire text 
needs to be checked and proofread by professionals from English speaking countries.

View manuscript.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Introduction: The introduction could be more concise. Some paragraphs are lengthy and repetitive (e.g., the discussion of sustainable entrepreneurship's importance is reiterated).
Measurement of Variables: While the constructs (e.g., economic empowerment, ICT quality) are well-defined, the paper could provide more detail on the survey items or scales used (e.g., Likert scale ranges, sample questions).

Figures 1 and 2 (conceptual models) are referenced but not included in the provided text. Ensure they are clearly labeled in the final manuscript.

Comments on the Quality of English Language

English language needs to be reviewed by a specialist

Author Response

Introduction

Restructured.

Figures 1 and 2

Are already cited in the text

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Please see the attached document.

Comments for author File: Comments.pdf

Author Response

Simultaneidad y Endogeneidad
We are aware that variables such as economic empowerment and sustainable entrepreneurship could be jointly determined, which could generate simultaneity biases and bidirectional causality. This situation may distort the causal relationships estimated by our structural equation modeling (SEM) approach. In order to address this issue and reduce potential endogeneity concerns, we conducted complementary analyses using two-stage least squares estimation (2SLS) as a robustness test.

For this purpose, we selected instrumental variables (IV) that affect economic empowerment but that, at the same time, do not present a direct correlation with sustainable entrepreneurship performance. Among the instruments chosen are community participation and perception of ICT quality, whose validity was supported both conceptually and empirically, by means of di-agnostic tests of relevance and exclusion restrictions. The results obtained from the 2SLS estimation were compared with those of the SEM model to assess the robustness and consistency of the estimated relationships. This comparison revealed no significant differences in the magnitude or significance of the parameters, which reinforces the confidence that possible simultaneity biases did not substantially affect the SEM results.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

1. ICT is suggested to be explained when it first appears in the abstract.
2. In line 48, 'In fact, the research by [3] is particularly vital...' might want to check the APA style.
3. Reference citations are used to support the authors' perspective and rationale behind the study; however, many reference citations in the manuscript were cited as a single source of reference for a sentence, seemingly lacking a systematic and overall literature review of the subject matter. It is suggested to enrich the reference citation for a solid research background for the study.
4. Since this is an empirical study based on a previous paper, in line 59, "psychological dimension [4]. ....." However, there is a void of causal variable relationships in the paper [4] which can serve as a research framework for the present study.  The authors need to justify and strengthen the support for the present developed research framework.
5. For abbreviations in Tables, need to provide notes explaining their full names for easier understanding.

Comments on the Quality of English Language

English quality is seemingly fine, but APA format is weird.

Author Response

1. ICT is suggested to be explained when it first appears in the abstract.

updated summary

1. In line 48, 'In fact, the research by [3] is particularly vital...' might want to check the APA style.

the style used by the publisher is MDPI (which resembles IEEE) and is not APA

1. Reference citations are used to support the authors' perspective and rationale behind the study; however, many reference citations in the manuscript were cited as a single source of reference for a sentence, seemingly lacking a systematic and overall literature review of the subject matter. It is suggested to enrich the reference citation for a solid research background for the study.

more authors were added

1. Since this is an empirical study based on a previous paper, in line 59, "psychological dimension [4]. ....." However, there is a void of causal variable relationships in the paper [4] which can serve as a research framework for the present study.  The authors need to justify and strengthen the support for the present developed research framework.

added justification

1. For abbreviations in Tables, need to provide notes explaining their full names for easier understanding.

added notes

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The reviewers raised the following questions, and the authors are invited to carefully consider.

1. Please indicate at the end of each section in the second section any deficiencies or additional research needed in the citation to demonstrate the innovation and scientific validity of this study. 2. Add the practicality and theoretical basis of each established mathematical model in the theoretical model section. 3. In the conclusion and discussion section, conduct a more in-depth analysis and discussion of the key conclusive data that emerged during the analysis process, and explain the reasons for their occurrence. 4. In the conclusion section, duplicate content is directly deleted, and the key points discovered in this study are explained and emphasized.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

1. Por favor indique al final de cada sección en la segunda sección cualquier deficiencia o investigación adicional necesaria en la cita para demostrar la innovación y validez científica de este estudio.

Enfoque y diseño de la investigación

Si bien los diseños no experimentales presentan limitaciones teóricas para inferir causalidad, la presentación de una estrategia de modelado de ecuaciones estructurales (SEM) mejora la validez interna del análisis al identificar constructos latentes y vías indirectas complejas. En el futuro, la investigación podría aprovechar este diseño en combinación con enfoques longitudinales o cuasiexperimentales para mejorar las afirmaciones causales.

Población y muestra

Aunque la muestra se limitó a 60 participantes por razones logísticas, esta limitación pone de manifiesto una brecha en la investigación empírica rural. Un mayor tamaño de muestra en futuras réplicas podría aumentar la generalización. Sin embargo, la muestra ofrece una oportunidad específica para una exploración a fondo y representa una valiosa prueba piloto para programas de empoderamiento de la RSU, ya existentes o potenciales, que deseen emularla.

Duración e Intervención del Programa

Una limitación de la intervención es su duración de solo seis meses, lo que significa que no se pueden capturar plenamente los efectos retardados en los resultados empresariales. Sin embargo, futuras investigaciones sobre empoderamiento deberían evaluar el impacto a largo plazo mediante evaluaciones de seguimiento. En cualquier caso, las fases estructuradas de esta intervención presentan un modelo replicable e innovador para la rápida implementación del empoderamiento.

Variables de estudio

Asimismo, cabe destacar que la operacionalización de estas variables en constructos latentes mediante escalas validadas fortalece el modelo SEM. Sin embargo, es recomendable validar algunos constructos de forma más exhaustiva, como la resiliencia, la alfabetización digital y la transformación de roles de género, principalmente en el Perú rural, para estudios de desarrollo de escalas en trabajos futuros.

Recopilación y análisis de datos

Dado que los datos autoinformados son significativos, los resultados de futuros estudios podrían mejorarse sustancialmente mediante la triangulación con indicadores objetivos de rendimiento (p. ej., niveles de ingresos, longevidad empresarial, entre otros). Esto no solo reducirá el sesgo común del método, sino que también aumentará su validez externa.

Consideraciones éticas

Aunque este estudio cumplió con las normas éticas, trabajos futuros podrían incluir consideraciones éticas más profundas, abordando enfoques éticos participativos que permitieran a los participantes definir criterios de evaluación e interpretar los hallazgos.

2. Agregue la practicidad y la base teórica de cada modelo matemático establecido en la sección del modelo teórico.

Fin del último párrafo de la Introducción

El modelo teórico presentado en este estudio integra un conjunto de vías causales que han sido validadas conceptual y empíricamente en la literatura. Los modelos directos de cada relación se basan en investigaciones previas de SEM en contextos de desarrollo rural [3,5,7], por ejemplo, y se adaptan aquí para el caso peruano. La calidad de las TIC hacia la vía de efectos de las TIC se basa en el modelo de aceptación de la tecnología y marcos más amplios de inclusión social y digital [6,13]. Los efectos de las TIC en los factores mediadores de la autonomía y la autoestima, por otro lado, se basan principalmente en la teoría de la integración social y la literatura sobre la utilidad percibida para la participación [1,7]. El papel de las TIC en el acceso de la comunidad a las oportunidades económicas se basa en la teoría rural y del valor, que sugiere que la organización colectiva mejora el acceso a los mercados y recursos [4,10]. La inclusión de factores como la autonomía, la autoestima y el contenido de las TIC se basa en la teoría de la psicología del empoderamiento [8] y la literatura sobre desarrollo participativo [2,11], mientras que la mediación de los factores de empoderamiento económico se fundamenta en modelos de empoderamiento basados ​​en recursos [5,8,27]. La validez matemática de este modelo se garantiza mediante el análisis de factores de rendimiento (AFC) y el uso del software SmartPLS, que permite estimaciones robustas de los efectos directos, indirectos y mediados, incluso en muestras pequeñas. En consecuencia, la estructura teórica y numérica del modelo representa un marco coherente y empíricamente validado para explicar cómo los programas de RSU impactan en el emprendimiento femenino sostenible en entornos rurales.

Subsección: Recopilación y análisis de datos

Además, el uso de SEM se sustenta en la teoría del empoderamiento y la teoría de sistemas, así como en software, lo que permite modelar relaciones complejas entre diversos constructos latentes, como el empoderamiento económico y la sostenibilidad [4,5,8]. El uso de PLS-SEM se justifica matemáticamente con base en muestras pequeñas y distribuciones no normales. Este método proporciona una estimación robusta mediante el método bootstrap, con 5000 remuestreos, y la posibilidad de estimar efectos directos e indirectos. Por lo tanto, esta convergencia metodológica corrobora la validez y la novedad del enfoque analítico de la presente investigación.

3. En la sección de conclusiones y discusión, realice un análisis y una discusión más profundos de los datos concluyentes clave que surgieron durante el proceso de análisis y explique las razones de su aparición.

Discusión

Un hallazgo aún más significativo fue el aumento estadísticamente significativo en la trayectoria de accesibilidad a las TIC a calidad de las TIC, con un cambio de β = 0,025 (p = 0,795) antes del programa a β = 0,347, p = 0,001 después del programa. Estos datos indican que el programa USR no solo fue eficaz en la provisión de TIC, sino que también reforzó la percepción de los miembros sobre la utilidad de las TIC y su capacidad informativa, lo que implica una mayor madurez digital. Este hallazgo respalda plenamente los hallazgos de [4] de que la capacitación puede transformar el acceso a las TIC de una barrera a un trampolín hacia la inclusión social. Mientras tanto, la trayectoria de calidad a impacto de las TIC fue sólida y alta en todo momento. β = 0,863 antes; β = 0,911 después, lo que sugiere un ciclo de retroalimentación. Esto puede explicar, a través del enfoque de los materiales en cada contenido y usabilidad, una opinión personal de [7] de que la capacitación adaptada a la comunidad mejora la percepción de las TIC. La trayectoria desde la participación comunitaria hasta el acceso a oportunidades económicas persistió y se fortaleció de β = 0,735 a β = 0,775, lo que sugiere resiliencia estructural de las redes comunitarias en comunidades rurales. Este marco se ha extraído de la intervención a través de redes estructurales como la solidaridad comunitaria y patrones de liderazgo establecidos que facilitaron el acceso a recursos, como se observó en entornos similares por [1]. Curiosamente, aunque la autoestima no fue significativa, el coeficiente β aumentó de 0,076 a 0,309, lo que sugiere que hay una dinámica psicosocial que emerge más tarde. Es probable que esto indique efectos latentes que requieren más tiempo para materializarse, quizás en línea con el umbral psicosocial propuesto por nosotros. Este fenómeno ha sido teorizado previamente en estudios [11] como efectos de empoderamiento con un desfase temporal en el riesgo entre las mujeres en áreas rurales extremadamente remotas. En general, estos datos convergentes confirman la hipótesis de que las intervenciones de RSU, cuando son multidimensionales y basadas en la comunidad, no solo producen dividendos técnicos rápidos sino que también desencadenan cambios estructurales a largo plazo.

Conclusiones

Ver manuscrito.

4. En la sección de conclusiones, se elimina directamente el contenido duplicado y se explican y enfatizan los puntos clave descubiertos en este estudio.

En resumen, los resultados de esta investigación proporcionan evidencia empírica sólida de que los programas de RDU tienen el potencial de catalizar transformaciones multidimensionales y sostenibles en el emprendimiento de las mujeres rurales. Si bien el enfoque de modelos de ecuaciones estructurales revela que la RDU puede ser no solo una intervención que mejora las capacidades técnicas, sino que también estimula cambios psicosociales y económicos más profundos. Por otro lado, uno de los hallazgos más concluyentes es el aumento estadísticamente significativo de la relación entre la calidad de las TIC y su impacto (β = 0,863 a β = 0,911, p < 0,001), lo que destaca una transición desde la adopción de tecnologías básicas hacia un uso más productivo de TIC de mayor calidad.

Cabe destacar que el estudio muestra un patrón de desarrollo adaptativo en tres fases: (1) acceso a la tecnología, (2) calidad del contenido técnico y (3) relevancia y contextualización del contenido. A diferencia de los supuestos estáticos y lineales que han dominado la literatura, el modelo postula un crecimiento integral y dinámico. El surgimiento de la autonomía en el contenido y las TIC como impulsores más poderosos que el acceso respalda la creencia de que los grupos con mayor experiencia exigen una intervención más contextualizada y relevante en diversas áreas de la infraestructura.

Además, se aclaran los roles del empoderamiento económico como mediador e impulsor, conectando las dimensiones técnicas, sociales y psicológicas. El fortalecimiento de la relación entre la participación comunitaria y las oportunidades económicas (β = 0,735 a β = 0,775, p < 0,001), combinado con el aumento de la autoestima (β = 0,076 a β = 0,309), indica el surgimiento de nuevos sistemas de apalancamiento psicosocial para el emprendimiento sostenible. Por lo tanto, una vez más, estos hallazgos respaldan la necesidad de iniciativas de RSU que se desarrollen en su implementación y que impulsen el desarrollo de capacidades comunitarias, las habilidades empresariales y el empoderamiento personal.

Sin embargo, el presente estudio presenta limitaciones. Además del pequeño tamaño de la muestra, la limitación geográfica a Chepén y Pacasmayo y el estudio transversal limitan la generalización y la inferencia a largo plazo. Asimismo, la dependencia de medidas autoinformadas podría haber introducido un sesgo metodológico común. Las futuras investigaciones deberían adoptar enfoques longitudinales y mixtos, validar el modelo propuesto en diferentes contextos, profundizar en el crecimiento psicosocial y evaluar cómo las tecnologías emergentes pueden beneficiar el emprendimiento sostenible entre las mujeres rurales.

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

The authors have addressed reviewer's cocerns snd is ready to proceed for publication 

Author Response

The authors have addressed reviewer's cocerns snd is ready to proceed for publication.

Thank you very much for everything!