Using regression and XGBoost analysis to explore expertise related to united nations sustainable development goals and research impact metrics among nursing faculty: a retrospective cross-sectional machine learning study
File(s)
Accepted version
Author(s)
Ruksakulpiwat, Suebsarn
Thongking, Witchuda
Benjasirisan, Chitchanok
Phianhasin, Lalipat
Praha, Nattaya
Type
Journal Article
Abstract
Background
The United Nations Sustainable Development Goals (SDGs) offer a comprehensive global framework for promoting health, equity and sustainability. Whereas alignment with the SDGs is increasingly encouraged in academic institutions, the extent to which faculty
expertise in SDGs influences traditional research impact metrics remains insufficiently explored.
Objective
To investigate the relationship between nursing faculty expertise in SDGs and research
impact metrics.
Methods
A retrospective cross-sectional design was employed using data from 121 nursing faculty members at Mahidol University, Thailand. Information on SDG-related expertise and research performance was obtained from the Mahidol University Research Excellence Database (MUREX) and Scopus. Descriptive statistics, Pearson’s correlation, and multiple linear regression analyses were used to examine associations between SDG expertise, academic experience, and research impact metrics, including H-index, citation count, and research output. Extreme Gradient Boosting (XGBoost) and Synthetic Minority Over-sampling Technique (SMOTE) were applied to improve predictive modelling and address class imbalance.
Results
Faculty members with greater expertise in SDGs demonstrated significantly higher research impact metrics. SDG expertise significantly predicted H-Index (β=0.65, p<0.001), total citations (β=31.77, p=0.004), and total research output (β=2.41, p<0.001). Research experience was also a significant predictor of research impact. XGBoost outperformed traditional regression models, identifying SDG expertise and international collaboration as the strongest predictors of research impact. Faculty aligned with SDG13 (Climate Action) had a high percentage of top-cited publications, suggesting
an intersection between research visibility and global sustainability priorities. Certain SDGs (e.g., SDG12: Responsible Consumption and SDG15: Life on Land) were
underrepresented, indicating potential gaps in nursing research.
Conclusion
SDG expertise is a key determinant of academic impact, reinforcing the need for greater institutional support for SDG-aligned research. Findings suggest that interdisciplinary collaboration and engagement with broader sustainability challenges may enhance faculty research visibility. Future research should explore longitudinal trends and policy implications for integrating SDGs into faculty assessment frameworks.
The United Nations Sustainable Development Goals (SDGs) offer a comprehensive global framework for promoting health, equity and sustainability. Whereas alignment with the SDGs is increasingly encouraged in academic institutions, the extent to which faculty
expertise in SDGs influences traditional research impact metrics remains insufficiently explored.
Objective
To investigate the relationship between nursing faculty expertise in SDGs and research
impact metrics.
Methods
A retrospective cross-sectional design was employed using data from 121 nursing faculty members at Mahidol University, Thailand. Information on SDG-related expertise and research performance was obtained from the Mahidol University Research Excellence Database (MUREX) and Scopus. Descriptive statistics, Pearson’s correlation, and multiple linear regression analyses were used to examine associations between SDG expertise, academic experience, and research impact metrics, including H-index, citation count, and research output. Extreme Gradient Boosting (XGBoost) and Synthetic Minority Over-sampling Technique (SMOTE) were applied to improve predictive modelling and address class imbalance.
Results
Faculty members with greater expertise in SDGs demonstrated significantly higher research impact metrics. SDG expertise significantly predicted H-Index (β=0.65, p<0.001), total citations (β=31.77, p=0.004), and total research output (β=2.41, p<0.001). Research experience was also a significant predictor of research impact. XGBoost outperformed traditional regression models, identifying SDG expertise and international collaboration as the strongest predictors of research impact. Faculty aligned with SDG13 (Climate Action) had a high percentage of top-cited publications, suggesting
an intersection between research visibility and global sustainability priorities. Certain SDGs (e.g., SDG12: Responsible Consumption and SDG15: Life on Land) were
underrepresented, indicating potential gaps in nursing research.
Conclusion
SDG expertise is a key determinant of academic impact, reinforcing the need for greater institutional support for SDG-aligned research. Findings suggest that interdisciplinary collaboration and engagement with broader sustainability challenges may enhance faculty research visibility. Future research should explore longitudinal trends and policy implications for integrating SDGs into faculty assessment frameworks.
Date Acceptance
2026-03-06
Citation
Journal of Nursing Management
ISSN
0966-0429
Publisher
Wiley
Journal / Book Title
Journal of Nursing Management
Copyright Statement
Copyright This paper is embargoed until publication. Once published the author’s accepted manuscript will be made available under a CC-BY License in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy).
License URL
Publication Status
Accepted
Date Publish Online
2026-03-06