Dependable data

The raw data on salaries referred to by Giles Dove of Stirling University and Roderick Floud of London Guildhall University (Letters, THES, June 9) were provided to the Association of University Teachers by the Higher Education Statistics Agency. The data provided by Hesa were in turn provided by the institutions themselves.

The AUT calculated the percentage difference between male and female staff in grade groupings provided by Hesa, and institutional rankings for each grade. These data were submitted to Hesa for scrutiny prior to publication, and it gave permission to publish the data.

Mr Dove said the Hesa/AUT figures were wrong because they were not the same as Stirling's data on median salaries for professors. That may well be the case, since the Hesa/AUT data were means, not medians. Hesa data clearly show that full-time female professors at Stirling in 1998-99 earned Pounds 43,928.20, while their male colleagues earned Pounds 54,545.64, some 24.2 per cent higher than the females. This difference was the highest in the UK. Also included in the professor category by Hesa were staff at pre-1992 institutions on research grade IV (none at Stirling in 1998-99, according to Hesa).

Mr Floud said the data did not take into account whether researchers were employed part time, but our report clearly stated that the data on researchers were for full-time staff on fixed-term contracts. So part-timers were excluded. The data did not control for age, but they did for grade, by grouping together staff on researcher A and B grades at post-1992 institutions and those on locally determined scales (none of the latter at LGU in 1998-99, according to Hesa). Senior researchers were included in the senior lecturer and professor categories. Full-time female researchers at LGU in 1998-99 earned Pounds 13,991.20 on average, while their male colleagues earned Pounds 20,026.88, 43.1 per cent higher than the females. This difference was the highest in the UK.

It is worth pointing out that the data provided by Hesa are only as accurate as the data provided by the institutions to Hesa. The AUT has made available to The THES the original spreadsheets with the Stirling and LGU data so that an independent party can check the validity of the figures. We fully defend our data: they were not "dodgy". It is surely now accepted across HE that it is discrimination that is "dodgy".

David Triesman. General secretary. Association of University Teachers.