Governance (v1.0)

Comprehensive Subnational Corruption Index (SCI)

The Comprehensive Subnational Corruption Index (SCI) provides a widely available standardized measure of overall corruption at the subnational level, combining both experienced petty corruption (SPCI) and perceived grand corruption (SGCI). It is constructed as the simple average of these two sub-indices, ensuring equal weighting between everyday bribery experiences and perceptions of systemic corruption. When either the SPCI or SGCI is missing, the missing value is estimated using a regression model that predicts one from the other, based on data from areas where both indices are available. This ensures broad subnational coverage while maintaining internal consistency.

Unlike the Baseline SCI, the Comprehensive SCI expands coverage through inter- and extrapolation, ensuring broader temporal and spatial availability. For countries with data available for multiple years, linear interpolation is used to estimate values in years without direct survey data. Additionally, for countries or years lacking survey-based corruption estimates, we extrapolate SCI values using national-level trends in the World Bank’s Control of Corruption Index (CCI). This process ensures the SCI maintains internal consistency while extending its coverage to a greater number of regions and years, making it particularly valuable for descriptive and policy-oriented research.

The SCI is directly comparable to leading national corruption indices, including the World Bank Control of Corruption Index (CCI) and Transparency International Corruption Perceptions Index (CPI). However, the SCI is based fully on individual perceptions and experiences, while the CCI and CPI partially rely on expert opinion. The SCI is also decomposable into its two components: petty and grand corruption. Furthermore, unlike these indices, the SCI provides subnational estimates, revealing within-country variation in corruption levels that national measures obscure. To enhance its usability, subnational versions of both the CCI and CPI, calculated by adjusting their national values based on the subnational SCI, are available at CorruptionRadar.

The index is scaled from 0 (most corrupt area in the dataset) to 100 (least corrupt area), with higher values indicating lower overall corruption. It is derived from 807 surveys covering 1,326,656 respondents across 1,473 subnational regions in 178 countries (1995–2022). To maximize cross-survey comparability, dimensions were harmonized, and missing values were imputed using chained regression models.

By offering a more complete picture of subnational corruption over time, the Comprehensive SCI is an invaluable tool for policymakers, researchers, and international organizations. It enables the identification of corruption hotspots, tracking corruption trends, and assessing the impact of governance reforms. While the Baseline SCI is best suited for explanatory research, the Comprehensive SCI is preferred for descriptive analyses, long-term monitoring, and within-year global comparisons.