This is the fourth year the Financial Times has ranked the full-time MBA programmes of the world's top business schools. The ranking provides a comprehensive assessment of the value of an MBA and the business schools that provide them. The criteria we have chosen enable comparison between schools globally.

This year, there have been two enhancements to the survey, with the application of specified auditing procedures to data supplied by 10 of the schools and a change in the way the final scores are calculated.

However, the position of a school in the table is still determined by its performance in three broad areas: the career progression accrued from the MBA - particularly its purchasing power in the marketplace - diversity of experience and the school's research qualities.

In the previous three rankings, the final score measured performance on a relative basis by giving 100 basis points to the top business school in each criterion and the other schools a proportion of this depending on their score.

This year, however, we have replaced this method in favour of using z-scores (as used in the inaugural executive MBA survey). These take into account not only the differences between each business school, but also the spread of scores between the top and bottom school in a single criterion.

A consequence of this is that some schools have seen slightly more exaggerated movements in rank position than would have occurred if we had continued to use the old method for calculating the final score.

We have also chosen not to present a final score. This is because z-scores produce final scores that are negative for almost half of the schools in the ranking, and could prove misleading.

In June last year, 149 schools were selected to take part in the MBA2002 rankings. The ranking is compiled from two questionnaires, one for the business schools and the other for the class which graduated three years previously - in this instance 1998 - and from an independent assessment of research.

Surveying graduates from 1998 allows us to chart the progress of graduates from before the MBA through to graduation and into the workplace. Questionnaires were sent to all alumni, who returned them by post or completed them over the internet.

In total, 21,342 questionnaires were sent out, of which 6,147 were returned, a response rate of 28.8 per cent. This is only 0.2 percentage points higher than last year but postal returns were down significantly on previous years, possibly because of tighter security within postal services worldwide.

Some of the questions on the questionnaires have been amended to improve the quality of the information collected, but the criteria for the ranking remain the same as last year.

In line with the changes to the calculation of the final score, we have adopted the use of z-scores in the calculation of the "international experience" and "research" criteria. As these criteria will also generate negative numbers, we have opted to display the results for each of these criteria as a ranking rather than as a points score.

The first eight criteria reflect the performance of the MBA programme, of which the salary data is the largest component. Overall, these eight criteria constitute 55 per cent of the total weighting.

The salary data are standardised (because they are reported in different currencies) by conversion to US dollars with PPP (purchasing power parity) exchange rates estimated by the World Bank. The PPP exchange rates take into account the differences in price levels between countries. We exclude extraordinarily high salaries and salary data from respondents employed in the non-profit sector and students.

The final salary figures published take into account, where available, salaries from the previous two surveys. The salary figure in the "Today" column is made up of 50 per cent of the average salary for graduates from the class of 1998, and 25 per cent each of salaries for the class of 1997 and 1996. For schools with only two years' data, the weighting is 60 per cent for 1998 and 40 per cent for 1997. The 1998 salary data alone is used for schools that are participating in the survey for the first time.

The salary data under the "weighted" column adjusts for the variation in salaries between different sectors. It is assumed, for example, that for the most recent sample each school has 27 per cent of alumni employed in finance/banking; 18 per cent in consultancy; 10 per cent in the industrial sector and 16 per cent in IT/telecommunications and the remaining 28 per cent in a consolidated "other" sector. This mirrors the overall response from alumni.

We do not calculate weighted salaries for small schools - those with fewer than 50 alumni questionnaire returns - because they do not have adequate responses in each sector.

The percentage increase is the increase in salary from the start of the MBA programme to the present day salary.

The value-for-money criterion is a short-term indicator that reflects the rate of return on the cost of the MBA after three years. We estimate a salary growth rate based on the average salary before the MBA, on graduation and at the time of sampling. We then divide the estimated income in the three years from graduation to today by the total cost of the course, including the opportunity cost of lost salary. The resulting figure gives the amount earned for every dollar spent on the course.

Career progression compares the level of responsibility and size of company in which alumni are employed from the start of the MBA to the present. A points system is used to measure the progression and the relative increase is calculated for each school.

The effectiveness of career services is measured by placement success. Alumni were asked to rate how useful they found their careers service in their job search.

The next eight criteria measure the diversity within a school among its faculty, board members and through the exposure and experience of the MBA student. These criteria reflect the experience of alumni while doing their MBA.

A points system is used to calculate the most internationally mobile alumni. Points are awarded to a school when alumni take up employment outside the country in which the business school is situated and who are subsequently employed in a third country.

International experience measures the degree to which alumni gain experience overseas through projects, study tours and exchange programmes.

The remaining criteria in the diversity section evaluate the percentage of international and women members on the school's advisory board, the percentage of international and women students, and language requirements.

The final three criteria measure the school's performance in research and account for 25 per cent of the total weighting. The research rating is based on the number of publications in 35 international and practitioner journals. The journals were chosen from a short-list of suggestions from the business schools, and the period over which we assess publications is January 1999 to the present. For each publication, points are awarded to the school where the faculty member is currently employed.

Additional research by Mark Barber and Ursula Milton