Perception errors

Illustration by Surendra.  

Before >I reported on the latest global Corruption Perception Index brought out by Transparency International, I was under the impression that the >world’s most widely used measure of corruption, while only a measure of perception and not of actual corruption, measured public perception. This, I found, is not the case.

To be clear, Transparency International is quite upfront about its sources. Essentially, it uses 12 surveys and assessments done by other organisations, and then standardises them to a scale of 0 to 100. In India’s case, nine of the 12 surveys are applicable. I looked up the websites each of them and their methodologies to get a sense of exactly whose perceptions we are talking about.

Just five of these 9 involve polls at all. The other four - the >Political Risk Services’ International Country Risk Guide, the >World Bank’s Country Policy and Institutional Assessment, the >Economist Intelligence Unit’s Country Risk Assessment and >IHS Global Insight’s Country Risk Ratings – all involve expert assessment, including in the Bank’s case, by its staff.

Of the five polls that do involve surveys, all but one are polls of business executives and/ or experts. So just one – the >World Justice Project’s Rule of Law Index – involves a poll of the general public, and for India this was just 1,047 people in Mumbai, Delhi and Bangalore. It was also conducted in 2013, while the new Corruption Perception Index says that the perception of corruption went down this year. (It also polled 24 experts per country.)

The remaining four surveys poll a very limited number of executives as well. >Bertelsmann Foundation’s Transformation Index polls two executives per country. The >IMD World Competitiveness Yearbook polls 4,300 business executives globally – I wasn’t able to find how many they polled in India. The >World Economic Forum’s Executive Opinion Survey polls 95 business executives per country. >Political and Economic Risk Consultancy’s Asian Intelligence polls 100 business executives per country.

Experts, a few hundred business executives, 1,047 members of the general public – these are essentially the people whose perceptions on corruption the Index is made up of.

Transparency International says so itself. “The CPI incorporates the perceptions of experts and business executives in India regarding the levels of public sector corruption in the country. Transparency International does not directly conduct any primary data analysis for computing the CPI and we rely upon these external data sources to calculate the index,” Santhosh Srinivasan, Transparency International’s Research Coordinator explained to me.

Is this a problem?

“From our research and analysis there is very close correlation between the perceptions of experts and the perceptions of the general public. In addition the CPI score also quite closely relates to the experiences of the citizens with regards to bribe payments. Therefore, it is our belief that the CPI provides an accurate reading of the general levels of public sector corruption in the country,” Mr. Srinivasan said.

>Critics have suggested that the CPI should be dropped entirely, and more direct information on corruption be gathered instead. But as you can imagine, direct information on corruption is rare. Until then, it will probably stand, but it’s useful knowing whose perception it reflects.

Our code of editorial values

This article is closed for comments.
Please Email the Editor

Printable version | Jun 15, 2021 6:59:36 AM |

Next Story