Transparency International unreliable!

SIR-I was dismayed by the Transparency International (TI) statistics in The New Vision which ranked Uganda the third most corrupt country in the world. What items were included in the corruption basket? The accuracy of survey statistics depends also on the sample size: What was the sample size? A good statistic estimator must have the following properties: 1. Unbiased: An estimator is said to be unbiased if its expected value is equal to the population parameter it estimates. Today Uganda to be ranked third most corrupt country in the world, considering all the transformations it has gone through, the in-streaming investment and the realisation of the government policies would alone show the TTs statistic is biased. I would say that there is a systematic deviation of the estimator from the parameter of interest, what is referred to as a bias. 2. Efficient: An estimator is said to be efficient if it has small variance (standard error). We say that one estimator is efficient relative to another. That is, if many other surveys were conducted, I doubt if their ranking would tend to rank Uganda third most corrupt country in world, rather than away from being the third. 3. Consistent: An estimator is said to be consistent if its probability of being close to the parameter it estimates increases as the sample size increases. Financial consideration always makes us take small sample size. The price is a poor estimator. That is why it is important to know the sample size in the corruption basket that TI took. 4. Sufficient: An estimator is said to be sufficient if it contains all the information in the data that the parameter estimates. I think it is mostly here that TI has the biggest ills. The same ills ranked Uganda in the late 1980s-early 1990s as the country most hit by HIV/AIDS in the world. Probably, this was not true, and was because of Uganda's openness on HIV/AIDS, thanks to President Yoweri Museveni. The same may be attributed to the current government's anti-corruption crusade. It (anti-corruption crusade) has left the Uganda books so wide open that all the data to depict corruption is there for all to see. How about other countries? Is there sufficient information from which to estimate correctly the corruption statistic estimators? So, with the available data, TI has used the classical approach (objective probability)_to compute the reported (The New Vision, June 29) statistic. TI should combine the prior information of Uganda's openness and all the information in the data to give us a good estimator of corruption. Julius Babyetsiza Director JB Computing Ltd, Kampala