The CPI ranking 1995

Many countries scored similarly in the 1995 and in the 1996 Ranking. Comparing the 1995 and the 1996 ranking, a lower score indicates a worse performance, whereas a higher number indicates a perception of improvements. However, keep in mind the following reminders:

• Differences in the two scores which are close to zero are not capable of indicating a significant change.
• In case a different score is accompanied with a high variance, this may be due to a stochastic outlier (a misperception by one source).
• A different score does not necessarily indicate a different performance. It may also be due to a larger set of sources and a more precise measurement of the subjective evaluations provided by our contributors.
• A changing performance may be due to actual regime shifts and a trend towards increasing or decreasing corruption over time. However, conclusions in this regard should be based on the comparison of the historical data.

The ranking for 1995 presents a list of 41 countries. The index has been prepared as a "poll of polls" by using seven surveys. Into the ranking three surveys (1992-1994) have been included from the World Competitiveness Report, Institute for Management Development, Lausanne. These results were "drawn from a unique survey of more than 2800 business executives worldwide". A subjective evaluation has been obtained by asking for the degree of "improper practices" in the respective home countries. Three more surveys (1992-1994) for Asia have been added from Political & Economic Risk Consultancy Ltd., Hong Kong. These surveys evaluate the degree of corruption by questioning more than seventy senior asian banks executives every year. They were asked to "rate on a scale of zero to 10 what they perceive to be the level of corruption of the countries of Asia with which they have first hand experience." Finally, the 1980 survey from Business International, New York, was added to the list. This survey rests on evaluations made by an unspecified number of BI Correspondents. Though it is out of date, the data appeared to be useful to us. Its validity is shown also by the fact, that it is highly correlated to the other rankings. Keep in mind, that all surveys measure corruption by gathering data about the subjective perception of corruption. This is due to the fact that objective data about corruption cannot be obtained. Note also, that the index does not necessarily reflect our personal opinion.

The following list contains three figures for every country. The first number indicates the average score in the surveys. For some data the scale had to be adjusted but we avoided any further manipulation. A "10" indicates a perfectly clean country whereas a "0" refers to a country where business transactions are entirely penetrated by corruption involving immense sums of kickbacks, extortion, fraud etc. The second number is the amount of surveys in which the particular country has been included (2-7). Note that countries which have only been ranked in a single survey have not been included in our ranking. The third number is the variance of the rankings. A high number indicates a high degree of deviating opinions. A variance of 0.01 for Denmark (9.32) indicates that the majority of thescores value will range between 9.22 to 9.42 (take the square root of the variance and add (subtract) the value for the upper (lower) value for a 68% interval). For South Africa, on the other hand, the high variance of 2.35 indicates rankings between 4.1 and 7.1. Apparently, the evaluation is quite diversified in this case.