Does the Evaluator Make a Difference? Measurement Validity in Corruption Research
IPSA Committee on Concepts and Methods Working Paper Series # 48 (2011)
33 Pages Posted: 15 Aug 2014
Date Written: 2011
This paper presents several tests of the validity of data on corruption using the full range of data sets employed in cross-national corruption research. First, we show that different classes of indicators of corruption, a distinction based on who evaluates a country’s level of corruption, rely on different standards to assess corruption and that the difference in standards is not consistent across countries. Second, we show that the problem with indicators of corruption is imported into Transparency International’s Corruption Perception Index (CPI) and the World Bank’s Control of Corruption Index (CCI). The disregard of systematic differences among indicators in the selection of indicators used in these indices, and in the formal weighting of these indicators, induces bias that undermines the validity of the CPI and the CCI. The implications of this assessment for the analysis and production of data on corruption, and other social-science concepts, are spelled out.
Suggested Citation: Suggested Citation