Predicting the status of 35 sustainable development goal indicators in Indian villages: a semi-supervised machine learning approach for precision public policy
Cross-sectional modeling study reveals wide variation in 35 development indicators across 597,603 Indian villages, indicating district averages mask critical local policy needs.
Key Points
To estimate village-level prevalence for 35 Sustainable Development Goal indicators across India in 2021 and evaluate required rates of improvement to achieve targets by 2030.
Linked cross-sectional data from the National Family Health Survey 2019–21 with the 2011 Indian Census.
Employed multilevel modeling to compute cluster-level values and applied a semi-supervised machine learning framework to project estimates for 597,603 villages.
Substantial heterogeneity was observed across all indicators; zero villages met 2021 targets for health insurance coverage, basic services, women's bank accounts, or internet use, whereas 99.9% achieved the adolescent pregnancy target.
The highest mean Required Rates of Improvement were needed for basic service access (7.61 percentage points/year), female health insurance (6.74), male health insurance (6.39), and clean cooking fuel (6.05).