Answers are grounded in this paper and cite their sources.
Composite indices are widely used to monitor subnational progress toward the Sustainable Development Goals (SDGs), yet their results can be sensitive to methodological choices, particularly indicator weighting. This raises concerns about their reliability for sustainability assessment and policy guidance, especially when robustness is not systematically evaluated. To address this issue, this study examines how alternative weighting schemes influence subnational SDG assessment through a framework that integrates multiple weighting approaches with explicit robustness analysis. The framework is applied to a harmonized dataset covering social, economic, environmental, and nutrition dimensions across Morocco’s 75 provinces. Composite indices are constructed using three alternative weighting approaches, namely equal weighting, entropy weighting, and principal component analysis (PCA), and evaluated using complementary robustness diagnostics. The results reveal a clear spatial structure of sustainability, with higher-performing provinces concentrated in metropolitan and coastal areas and lower-performing provinces located in interior and peripheral regions. These territorial patterns remain consistent across weighting schemes. However, intermediate-ranking provinces exhibit greater sensitivity to weighting assumptions, reflecting more heterogeneous development profiles and increased uncertainty in interpreting territorial performance. The findings indicate that sensitivity to weighting schemes reflects both methodological uncertainty and underlying differences in territorial development profiles. Robustness analysis therefore contributes not only to the validation of composite indices but also to the interpretation of territorial disparities. The study highlights the value of integrating robustness analysis into subnational SDG assessment and supports a transparent and context-aware use of composite indicators in sustainability analysis and policy planning.
No takes yet. Share an insight, caveat, or question.
Tayi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: