Assessing corruption: an analytical review of corruption measurement and its problems - perception, error and utility

Are corruption indicators reliable?

By: Edmund A
Published by: Development Gateway, 2007
Via: Eldis

This paper analyses how corruption is measured through an analytical review of major recognised corruption indicators. It examines the disparities between subjective and objective indicators, differentiating between large and pure objective indicators. In particular, the paper discusses the strengths and weaknesses of second-generation and aggregate indicators including:

• the Corruption Perception Index (CPI)
• the Business Environment and Enterprise Performance Survey (BEEPS)
• the World Governance Indicators (WGI)

The paper argues that there are three main problems with current indicators:

• the perception problem: indicators based on perception are too often interpreted as “real” levels of corruption. Moreover, the complex statistical constructions of modern aggregate indicators can easily create an illusion of quantitative sophistication that leads to interpret them as actual corruption indicators
  
• the error problem: because indicators use perception they include large margins of error due to the error relative to the specific concept that is expected to measure, and the imperfection of any proxy for such a concept
  
• the utility problem: corruption indicators have been largely criticised for offering too broad corruption assessments, difficult to convert into concrete anti-corruption efforts

The paper concludes corruption assessment has proved to be a complicated task subject to several difficulties such as the lack of objective data, the error measurement both endogenous and exogenous to corruption, and the complexity to build effective bridges from measurement to the solution.

(http://topics.developmentgateway.org/governance/rc/filedownload.do?itemId=1114356)