Agreed that six people is too small a base... but TI being "better" is little consolation!
*http://bit.ly/e777xW* * * ... it is unclear what exactly the CPI [Corruption Perception Index] is measuring, when sources which measure such different aspects of corruption are averaged together. It is a bit like adding, or in the CPI's case averaging, apples and oranges. To give a rough example, suppose that in city A there were 5 murders and 95 incidents of shoplifting, whereas in city B, there were 95 murders and 5 incidents of shoplifting. The size of the population is the same in both cities. Then, the total crime rate is the same in the two cities. But no one would venture to say that they are equally safe cities to live in. This is an exaggerated example of what happens in the CPI, where grand embezzlement and petty corruption are treated as the same entity.... ... According to PERC's ratings, Japan became almost 2q.4 times as corrupt (an increase of 138%) between 1996 and 1997. China, India and Malaysia also jump up almost two points each between 1998 and 1999. The reason why one might doubt the reliability of these numbers is that corruption is deep-rooted in a country, the result of a combination of factors, including weak institutions (such as the judiciary and property rights), over-regulation, entrenched bureaucray, and an under-developed civil society and media. These factors are unlikely to change so dramatically in a one-year period. That the scores display such unrealistic and dramatic variations is particularly troublesome for the [Corruption Perception Index].... Frederick Noronha :: +91-9822122436 :: +91-832-2409490 On 21 February 2011 18:34, Santosh Helekar <[email protected]> wrote: > > > The surveys conducted by Transparency International are more scientific and > better at assessing the level of corruption at any given time than talking > to 6 friends. >
