Rethinking Sustainability Indicators for the Extractive Industry: A Multitechnique Statistical Approach

ABSTRACT The extractive industry faces increasing pressure to report sustainability performance measured through sustainability indicators (SI). Comprehensive sustainability reporting frameworks like the Global Reporting Initiative (GRI) contain nearly a hundred indicators that create reporting complexity and reduce comparability. Although existing research emphasizes context‐specific SI for emerging markets, few empirical studies identify relevant context‐specific SI. This study empirically derives a manageable set of SI for India's extractive industry from GRI's G4 framework. Data from reporting companies over 9 years were analyzed using three statistical techniques: multiple correspondence analysis, nonnegative matrix factorization, and cosine similarity analysis across economic, environmental, and social pillars. The analysis identified 27 SI (one‐third of the original list) that maintained parity with industry leaders. Back‐testing validated that no information was lost during reduction. This approach is scalable and replicable for other industries and geographies, making it a useful tool for sustainability researchers and practitioners.