The emergence of artificial intelligence (AI) technology has led to transformative advancements across various industries, particularly in innovative sectors, creating both new challenges and opportunities. This study systematically examines the impact of AI technological innovation on global entrepreneurial activities. The basic theoretical framework is constructed using entrepreneurial factor theory and external enabler framework, and a two-way fixed effects model along with a mediation effect model are employed to study the impact of AI technological innovation on corporate entrepreneurial activities in 52 countries and regions from 2002 to 2023, spanning a total of 22 years. The results reveal that AI technological innovation significantly promotes entrepreneurial activities, particularly opportunity-driven ventures. Notably, the impact is especially significant for entrepreneurs aged 18–34 and those with a higher level of education. Additionally, AI technological innovation positively influences entrepreneurial activities through two paths: entrepreneurship education and AI investment. This study provides valuable insights for countries to effectively leverage AI technology, thereby enhancing entrepreneurial vitality and promoting economic resilience.
{"title":"The impact of artificial intelligence technological innovation on global entrepreneurial activities","authors":"Yuqi Tian , Xiaowen Wang , Nanxu Chen , Zhenhua Zhang","doi":"10.1016/j.seps.2025.102381","DOIUrl":"10.1016/j.seps.2025.102381","url":null,"abstract":"<div><div>The emergence of artificial intelligence (AI) technology has led to transformative advancements across various industries, particularly in innovative sectors, creating both new challenges and opportunities. This study systematically examines the impact of AI technological innovation on global entrepreneurial activities. The basic theoretical framework is constructed using entrepreneurial factor theory and external enabler framework, and a two-way fixed effects model along with a mediation effect model are employed to study the impact of AI technological innovation on corporate entrepreneurial activities in 52 countries and regions from 2002 to 2023, spanning a total of 22 years. The results reveal that AI technological innovation significantly promotes entrepreneurial activities, particularly opportunity-driven ventures. Notably, the impact is especially significant for entrepreneurs aged 18–34 and those with a higher level of education. Additionally, AI technological innovation positively influences entrepreneurial activities through two paths: entrepreneurship education and AI investment. This study provides valuable insights for countries to effectively leverage AI technology, thereby enhancing entrepreneurial vitality and promoting economic resilience.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102381"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145569016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-10-16DOI: 10.1016/j.seps.2025.102354
Ali Younes , Mohamed O. Abu Ghazala , Tamer A. Al-Sabbagh , Hamdy N. Eid , Mohamed A. El-Shenawy
The strategic siting of ammunition depots is a critical component of national safety and risk mitigation planning. Proper location selection minimizes potential threats to public safety and reduces the socio-economic impact of hazardous events. This study develops a spatial decision support model to identify optimal depot locations in Egypt, a developing country facing complex security and environmental challenges. The methodology integrates Geographic Information Systems (GIS) with Multi-Criteria Decision-Making (MCDM), specifically employing the Ordinal Priority Approach (OPA) to derive weights based on expert input. A total of seven evaluation criteria and seventeen spatial constraints were selected and validated by twelve subject-matter experts through structured surveys. The model also includes a sensitivity analysis to evaluate the robustness of outcomes under parameter variability. Results indicate that 44.36 % of Egypt's land area is suitable for ammunition depot siting, with 4.54 % highly suitable, 83.45 % moderately suitable, and 12.01 % marginally suitable. North Sinai and Matrouh governorates emerged as the most favorable regions. Ten potential locations are proposed within the highly suitable areas, offering actionable insights for defense infrastructure planning. The findings support evidence-based policymaking in the defense sector and demonstrate the value of spatial decision tools in optimizing sensitive facility siting within socio-economic and environmental frameworks.
{"title":"Spatial suitability and facility location planning for ammunition depots in Egypt: An integrated GIS-based MCDM approach","authors":"Ali Younes , Mohamed O. Abu Ghazala , Tamer A. Al-Sabbagh , Hamdy N. Eid , Mohamed A. El-Shenawy","doi":"10.1016/j.seps.2025.102354","DOIUrl":"10.1016/j.seps.2025.102354","url":null,"abstract":"<div><div>The strategic siting of ammunition depots is a critical component of national safety and risk mitigation planning. Proper location selection minimizes potential threats to public safety and reduces the socio-economic impact of hazardous events. This study develops a spatial decision support model to identify optimal depot locations in Egypt, a developing country facing complex security and environmental challenges. The methodology integrates Geographic Information Systems (GIS) with Multi-Criteria Decision-Making (MCDM), specifically employing the Ordinal Priority Approach (OPA) to derive weights based on expert input. A total of seven evaluation criteria and seventeen spatial constraints were selected and validated by twelve subject-matter experts through structured surveys. The model also includes a sensitivity analysis to evaluate the robustness of outcomes under parameter variability. Results indicate that 44.36 % of Egypt's land area is suitable for ammunition depot siting, with 4.54 % highly suitable, 83.45 % moderately suitable, and 12.01 % marginally suitable. North Sinai and Matrouh governorates emerged as the most favorable regions. Ten potential locations are proposed within the highly suitable areas, offering actionable insights for defense infrastructure planning. The findings support evidence-based policymaking in the defense sector and demonstrate the value of spatial decision tools in optimizing sensitive facility siting within socio-economic and environmental frameworks.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102354"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145365487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-10-14DOI: 10.1016/j.seps.2025.102351
Zhonghua Cheng , Guang Yang
As the core driver of the new round of technological innovation, artificial intelligence (AI) is profoundly transforming the governance models of enterprises, which may thus exert a significant impact on corporate ESG greenwashing. Accordingly, this research analyzes the listed enterprises in the A-share market from 2010 to 2023, employing a double machine learning (DML) model to examine the impact of AI on corporate ESG greenwashing. The findings reveal that: (1) AI significantly inhibits corporate ESG greenwashing. This conclusion remains robust after a series of tests, including model re-specification, variable substitution, and checks for endogeneity issues. (2) AI inhibits corporate ESG greenwashing by improving regulatory efficiency and reducing inefficient investments. (3) The AI's suppressive impact on corporate ESG greenwashing exhibits heterogeneous effects across corporate characteristics, particularly pronounced in state-owned enterprises, polluting enterprises, and large enterprises.
{"title":"The impact of artificial intelligence on corporate ESG greenwashing","authors":"Zhonghua Cheng , Guang Yang","doi":"10.1016/j.seps.2025.102351","DOIUrl":"10.1016/j.seps.2025.102351","url":null,"abstract":"<div><div>As the core driver of the new round of technological innovation, artificial intelligence (AI) is profoundly transforming the governance models of enterprises, which may thus exert a significant impact on corporate ESG greenwashing. Accordingly, this research analyzes the listed enterprises in the A-share market from 2010 to 2023, employing a double machine learning (DML) model to examine the impact of AI on corporate ESG greenwashing. The findings reveal that: (1) AI significantly inhibits corporate ESG greenwashing. This conclusion remains robust after a series of tests, including model re-specification, variable substitution, and checks for endogeneity issues. (2) AI inhibits corporate ESG greenwashing by improving regulatory efficiency and reducing inefficient investments. (3) The AI's suppressive impact on corporate ESG greenwashing exhibits heterogeneous effects across corporate characteristics, particularly pronounced in state-owned enterprises, polluting enterprises, and large enterprises.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102351"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145365491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-09-25DOI: 10.1016/j.seps.2025.102337
Simona Ferraro , Kaire Põder , Triin Lauri
This paper investigates how inclusive education reforms intersect with parental choice to influence school efficiency in Estonia - a system that is formally comprehensive, but increasingly selective in practice, leading to quasi-market dynamics. Applying a two-stage double-bootstrap Data Envelopment Analysis (DEA) on post-pandemic data from over 300 lower secondary schools, we assess how non-discretionary student characteristics (environmental variables), particularly special educational needs (SEN), parental income and immigration background, affect school-level efficiency. Our findings show that higher proportions of SEN students and students from low-income families are systematically associated with lower efficiency, especially in contexts where schools have no autonomy over admissions, such as neighbourhood schools. In contrast, oversubscribed or elite schools can afford to be selective, reinforcing reputational hierarchies and equity-harming quasi-market dynamics. By linking efficiency analysis with educational governance, we discuss how school market characteristics can easily jeopardise the inclusive education reform. Evidence shows that in a hybrid market, non-selective schools are worse positioned in terms of efficiency than selective schools.
{"title":"Inclusive education and parental choice: How student characteristics affect school efficiency","authors":"Simona Ferraro , Kaire Põder , Triin Lauri","doi":"10.1016/j.seps.2025.102337","DOIUrl":"10.1016/j.seps.2025.102337","url":null,"abstract":"<div><div>This paper investigates how inclusive education reforms intersect with parental choice to influence school efficiency in Estonia - a system that is formally comprehensive, but increasingly selective in practice, leading to quasi-market dynamics. Applying a two-stage double-bootstrap Data Envelopment Analysis (DEA) on post-pandemic data from over 300 lower secondary schools, we assess how non-discretionary student characteristics (environmental variables), particularly special educational needs (SEN), parental income and immigration background, affect school-level efficiency. Our findings show that higher proportions of SEN students and students from low-income families are systematically associated with lower efficiency, especially in contexts where schools have no autonomy over admissions, such as neighbourhood schools. In contrast, oversubscribed or elite schools can afford to be selective, reinforcing reputational hierarchies and equity-harming quasi-market dynamics. By linking efficiency analysis with educational governance, we discuss how school market characteristics can easily jeopardise the inclusive education reform. Evidence shows that in a hybrid market, non-selective schools are worse positioned in terms of efficiency than selective schools.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102337"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145419008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents the phenomenon referred to as Productive Vocation (PV), which is a methodological proposal of this study to measure and promote the economic development of territories. PV is defined as the existence of high productive sensitivity (elasticity) in response to a group of structural similarities (labor, productive capacity, functionality, specialization, entrepreneurship, among others) and interaction with other territories. The study focuses on the nodal centers of the Chilean macrozones Norte Chico and Patagonia from 2008 to 2018 as reference territories.
It was observed that the presence of this PV allows centers to act as true engines of development, with similarity and interaction fostering productive relationships. However, given the differences between the two macrozones, policies must be tailored to each territory, taking into account the variables that sensitize productive exchange.
Finally, it was observed how smaller centers (or intermediate cities) pressure and complement the larger ones to increase their sensitivities.
{"title":"Productive vocation - A methodological proposal for the analysis of territorial factors of localization and specialization as drivers of development","authors":"Sergio Soza-Amigo , Claudio Mancilla , Luz María Ferrada , Jorge Parada","doi":"10.1016/j.seps.2025.102357","DOIUrl":"10.1016/j.seps.2025.102357","url":null,"abstract":"<div><div>This paper presents the phenomenon referred to as Productive Vocation (PV), which is a methodological proposal of this study to measure and promote the economic development of territories. PV is defined as the existence of high productive sensitivity (elasticity) in response to a group of structural similarities (labor, productive capacity, functionality, specialization, entrepreneurship, among others) and interaction with other territories. The study focuses on the nodal centers of the Chilean macrozones Norte Chico and Patagonia from 2008 to 2018 as reference territories.</div><div>It was observed that the presence of this PV allows centers to act as true engines of development, with similarity and interaction fostering productive relationships. However, given the differences between the two macrozones, policies must be tailored to each territory, taking into account the variables that sensitize productive exchange.</div><div>Finally, it was observed how smaller centers (or intermediate cities) pressure and complement the larger ones to increase their sensitivities.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102357"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145323617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-10-16DOI: 10.1016/j.seps.2025.102356
Biyu Liu , Wanying Chen , Haidong Yang
Strict environmental regulations have led enterprises to emphasize the role of leasing in facilitating the return of used products for remanufacturing. How to make optimal operational decisions in a leasing-selling closed-loop supply chain (CLSC)? How does the carbon cap and trade policy (CCTP) affect operational decisions in a leasing and selling CLSC? To answer the above questions, this paper proposes four operational decision models for a leasing-selling CLSC under royalty/fixed-fee authorization strategy with and without the CCTP, respectively, are proposed. The optimal authorization strategies and operational decisions of entities in the CLSC and the win-win situation are analyzed. The results indicate: (1) There exists a win-win situation. Without the CCTP, it is influenced by the manufacturing cost per unit of new product (NP) used for leasing. When this cost is high the fixed-fee authorization strategy within a certain range authorization fee is a win-win option. Under the CCTP, besides the influence of the manufacturing cost per unit of NP used for leasing, the carbon emissions per unit of NP also affect the win-win situation. (2) When carbon emission savings are medium, the quantity of remanufactured products leased increases with the carbon trading price under the royalty authorization strategy while the opposite is true for the fixed-fee authorization strategy. (3) The increase in the remaining value of product at lease expiration makes the CLSC more environmentally friendly and enhances the emission reduction effect of the CCTP.
{"title":"Operational decision-making for a leasing-selling closed-loop supply chain with authorized remanufacturing under carbon cap and trade policy","authors":"Biyu Liu , Wanying Chen , Haidong Yang","doi":"10.1016/j.seps.2025.102356","DOIUrl":"10.1016/j.seps.2025.102356","url":null,"abstract":"<div><div>Strict environmental regulations have led enterprises to emphasize the role of leasing in facilitating the return of used products for remanufacturing. How to make optimal operational decisions in a leasing-selling closed-loop supply chain (CLSC)? How does the carbon cap and trade policy (CCTP) affect operational decisions in a leasing and selling CLSC? To answer the above questions, this paper proposes four operational decision models for a leasing-selling CLSC under royalty/fixed-fee authorization strategy with and without the CCTP, respectively, are proposed. The optimal authorization strategies and operational decisions of entities in the CLSC and the win-win situation are analyzed. The results indicate: (1) There exists a win-win situation. Without the CCTP, it is influenced by the manufacturing cost per unit of new product (NP) used for leasing. When this cost is high the fixed-fee authorization strategy within a certain range authorization fee is a win-win option. Under the CCTP, besides the influence of the manufacturing cost per unit of NP used for leasing, the carbon emissions per unit of NP also affect the win-win situation. (2) When carbon emission savings are medium, the quantity of remanufactured products leased increases with the carbon trading price under the royalty authorization strategy while the opposite is true for the fixed-fee authorization strategy. (3) The increase in the remaining value of product at lease expiration makes the CLSC more environmentally friendly and enhances the emission reduction effect of the CCTP.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102356"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145365551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-10-10DOI: 10.1016/j.seps.2025.102346
Elena Prodi , Chiara Pollio , Marco R. Di Tommaso , Manli Huang
This study explores the evolving landscape of industrial policy in a post-shock global context, where fostering the resilience of strategic assets has become crucial for both advanced and emerging economies. Within this framework, the paper develops a novel integrated methodology to assist decision-makers in addressing the complexities of industrial policy design and implementation. Specifically, our approach accounts for multiple dimensions of resilience, enabling a nuanced understanding of territorial responses across different domains. We apply this methodology to examine the resilience of Chinese provinces across both industrial and social dimensions in the aftermath of the COVID-19 shock. Our results show that industrial and social resilience do not always align; rather, they can diverge significantly, even within the same regional clusters. These findings open new theoretical avenues for analyzing the interaction—and at times tension—between different resilience domains within a single territorial context. Furthermore, we find that the relationship between manufacturing specialization and resilience may not be straightforward. Among provinces with higher manufacturing specialization, resilience tends to manifest strongly in either the social or industrial dimension, depending on the specific characteristics of their manufacturing sectors. Additionally, we show that provinces with greater investment in social policies were generally better equipped to absorb shocks. Overall, our findings offer valuable insights for policymakers by providing a more nuanced understanding of resilience and highlighting where targeted interventions may be necessary across different domains to support more balanced and inclusive recovery trajectories.
{"title":"Post-shock resilience and preference for manufacturing? A study on Chinese provinces","authors":"Elena Prodi , Chiara Pollio , Marco R. Di Tommaso , Manli Huang","doi":"10.1016/j.seps.2025.102346","DOIUrl":"10.1016/j.seps.2025.102346","url":null,"abstract":"<div><div>This study explores the evolving landscape of industrial policy in a post-shock global context, where fostering the resilience of strategic assets has become crucial for both advanced and emerging economies. Within this framework, the paper develops a novel integrated methodology to assist decision-makers in addressing the complexities of industrial policy design and implementation. Specifically, our approach accounts for multiple dimensions of resilience, enabling a nuanced understanding of territorial responses across different domains. We apply this methodology to examine the resilience of Chinese provinces across both industrial and social dimensions in the aftermath of the COVID-19 shock. Our results show that industrial and social resilience do not always align; rather, they can diverge significantly, even within the same regional clusters. These findings open new theoretical avenues for analyzing the interaction—and at times tension—between different resilience domains within a single territorial context. Furthermore, we find that the relationship between manufacturing specialization and resilience may not be straightforward. Among provinces with higher manufacturing specialization, resilience tends to manifest strongly in either the social or industrial dimension, depending on the specific characteristics of their manufacturing sectors. Additionally, we show that provinces with greater investment in social policies were generally better equipped to absorb shocks. Overall, our findings offer valuable insights for policymakers by providing a more nuanced understanding of resilience and highlighting where targeted interventions may be necessary across different domains to support more balanced and inclusive recovery trajectories.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102346"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145365490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-10-22DOI: 10.1016/j.seps.2025.102366
Mehmet Hakan Özdemir , Batin Latif Aylak , Celal Cakiroglu , Mahmut Bağcı
Gender inequality is acknowledged as a major hindrance to human development, evident in multiple social, political, economic, and cultural aspects. Therefore, identifying the factors contributing to gender inequality and quantifying them is crucial for enhancing societal progress. A new index, the gender inequality index (GII), was introduced in the 2010 Human Development Report to quantify and compare gender inequalities among different countries. Multiple indicators are used to calculate the GII, which involves complex analytical calculations. This study utilizes these indicators as input features to predict the GII using XGBoost, CatBoost, Extra Trees, LightGBM, Ridge, and Lasso regression models. These regressors are trained for predicting the GII as a function of maternal mortality ratio, adolescent birth rate, share of seats in parliament, female population with at least some secondary education, male population with at least some secondary education, female labour force participation rate, and male labour force participation rate. It is observed that XGBoost, CatBoost, Extra Trees and LightGBM predictors have score greater than 0.98, while the Ridge and Lasso regressors have score less than 0.90. The highest average accuracy is obtained by the CatBoost model while the XGBoost model has the greatest computational speed. Furthermore, the Shapley additive explanations methodology is utilized to detect the impact of different input features on the model predictions, and this information allows for more precise calculation of the GII. Thus, the proposed machine learning procedure enables both simplicity and flexibility for the GII prediction and provides more effective use of the GII.
{"title":"Prediction of the gender inequality index based on data-driven interpretable ensemble learning methods","authors":"Mehmet Hakan Özdemir , Batin Latif Aylak , Celal Cakiroglu , Mahmut Bağcı","doi":"10.1016/j.seps.2025.102366","DOIUrl":"10.1016/j.seps.2025.102366","url":null,"abstract":"<div><div>Gender inequality is acknowledged as a major hindrance to human development, evident in multiple social, political, economic, and cultural aspects. Therefore, identifying the factors contributing to gender inequality and quantifying them is crucial for enhancing societal progress. A new index, the gender inequality index (GII), was introduced in the 2010 Human Development Report to quantify and compare gender inequalities among different countries. Multiple indicators are used to calculate the GII, which involves complex analytical calculations. This study utilizes these indicators as input features to predict the GII using XGBoost, CatBoost, Extra Trees, LightGBM, Ridge, and Lasso regression models. These regressors are trained for predicting the GII as a function of maternal mortality ratio, adolescent birth rate, share of seats in parliament, female population with at least some secondary education, male population with at least some secondary education, female labour force participation rate, and male labour force participation rate. It is observed that XGBoost, CatBoost, Extra Trees and LightGBM predictors have <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> score greater than 0.98, while the Ridge and Lasso regressors have <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> score less than 0.90. The highest average accuracy is obtained by the CatBoost model while the XGBoost model has the greatest computational speed. Furthermore, the Shapley additive explanations methodology is utilized to detect the impact of different input features on the model predictions, and this information allows for more precise calculation of the GII. Thus, the proposed machine learning procedure enables both simplicity and flexibility for the GII prediction and provides more effective use of the GII.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102366"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145365493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Culture plays a central role in society. It supports inclusion, identity, and civic participation, while also contributing to economic activity and political engagement. Therefore, using appropriate tools to measure Cultural Engagement (CE) is crucial. In this paper, a set of composite indicators to measure cultural participation is developed, starting from individual participation in a range of cultural activities. Traditional methods—such as weighted and unweighted averages and Principal Component Analysis (PCA)—and a different approach based on Item Response Theory (IRT) are compared. These methods are examined in terms of interpretability and key features, looking across Italian regions and population subgroups, with attention to economic resources, education and gender. The analysis relies on nine waves (2014–2022) of the Aspects of Daily Life survey by the Italian National Institute of Statistics (ISTAT). While classical methods tend to be more intuitive and easier to read, the IRT-based approach offers distinct advantages, especially when analyzing individual cultural activities. Overall, the IRT approach enriches a body of literature that has, so far, offered limited tools for measuring CE.
{"title":"Which indicator best measures cultural engagement? A comparative analysis","authors":"Alessandro Gallo, Francesca Adele Giambona, Daniele Vignoli","doi":"10.1016/j.seps.2025.102379","DOIUrl":"10.1016/j.seps.2025.102379","url":null,"abstract":"<div><div>Culture plays a central role in society. It supports inclusion, identity, and civic participation, while also contributing to economic activity and political engagement. Therefore, using appropriate tools to measure Cultural Engagement (CE) is crucial. In this paper, a set of composite indicators to measure cultural participation is developed, starting from individual participation in a range of cultural activities. Traditional methods—such as weighted and unweighted averages and Principal Component Analysis (PCA)—and a different approach based on Item Response Theory (IRT) are compared. These methods are examined in terms of interpretability and key features, looking across Italian regions and population subgroups, with attention to economic resources, education and gender. The analysis relies on nine waves (2014–2022) of the <em>Aspects of Daily Life</em> survey by the Italian National Institute of Statistics (ISTAT). While classical methods tend to be more intuitive and easier to read, the IRT-based approach offers distinct advantages, especially when analyzing individual cultural activities. Overall, the IRT approach enriches a body of literature that has, so far, offered limited tools for measuring CE.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102379"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145569015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-11-01DOI: 10.1016/j.seps.2025.102339
Lütje Lange, Alexander Griffiths, Hans-Hennig von Grünberg
In this study, we examine how standard AI tools such as ChatGPT can be used for the structured analysis of large text corpora. To this end, we analysed 482 applications from a specific innovation funding program of the German Federal Ministry of Science using ChatGPT. Thanks to ChatGPT's ability to cluster projects based on their characteristics, complex data sets can be systematically explored and patterns recognized that would have remained hidden in a manual analysis. It turns out that cluster formation controlled in advance by the user via cluster definitions (using prompts), is in some cases more meaningful than the fully automated cluster formation of tools such as BERTopic. The analysis of the 482 funding applications provides detailed insights into the state of innovation in Germany: 83 % of the proposals dealt with topics related to digitalization and social innovation (half each), while the remaining 17 % dealt with sustainability issues. While 77 % of all project activities focus solely on the early concept phases, only 17 % of activities relate to the piloting and validation of applied ideas. Correlation analyses examine the relationships and potential connections between the clusters identified in different categories, in order to uncover patterns and dependencies in the innovation application data. For example, the correlation data can be used to determine the “age” of certain fields of innovation. The study also demonstrates the suitability of the method for classifications with external cluster definitions such as the UN Sustainable Development Goals (SDGs) or the EU program “Horizon Europe” to assess the suitability of research projects, with regard to specific frameworks. This could be particularly useful for scientific funding organizations.
{"title":"Introducing a novel AI-based text mining method illustrated through an analysis of German innovation proposals","authors":"Lütje Lange, Alexander Griffiths, Hans-Hennig von Grünberg","doi":"10.1016/j.seps.2025.102339","DOIUrl":"10.1016/j.seps.2025.102339","url":null,"abstract":"<div><div>In this study, we examine how standard AI tools such as ChatGPT can be used for the structured analysis of large text corpora. To this end, we analysed 482 applications from a specific innovation funding program of the German Federal Ministry of Science using ChatGPT. Thanks to ChatGPT's ability to cluster projects based on their characteristics, complex data sets can be systematically explored and patterns recognized that would have remained hidden in a manual analysis. It turns out that cluster formation controlled in advance by the user via cluster definitions (using prompts), is in some cases more meaningful than the fully automated cluster formation of tools such as BERTopic. The analysis of the 482 funding applications provides detailed insights into the state of innovation in Germany: 83 % of the proposals dealt with topics related to digitalization and social innovation (half each), while the remaining 17 % dealt with sustainability issues. While 77 % of all project activities focus solely on the early concept phases, only 17 % of activities relate to the piloting and validation of applied ideas. Correlation analyses examine the relationships and potential connections between the clusters identified in different categories, in order to uncover patterns and dependencies in the innovation application data. For example, the correlation data can be used to determine the “age” of certain fields of innovation. The study also demonstrates the suitability of the method for classifications with external cluster definitions such as the UN Sustainable Development Goals (SDGs) or the EU program “Horizon Europe” to assess the suitability of research projects, with regard to specific frameworks. This could be particularly useful for scientific funding organizations.</div></div>","PeriodicalId":22033,"journal":{"name":"Socio-economic Planning Sciences","volume":"103 ","pages":"Article 102339"},"PeriodicalIF":5.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145569095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号