Ranking Australian Universities: Controlling for Scope

Abstract

Rankings of universities that are based on aggregate measures ignore the discipline coverage within institutions. In this paper, for each of Australia's thirty‐seven public universities, performance in research and teaching in broad discipline groups is evaluated and then aggregated up to obtain whole‐of‐institution rankings. Relative staff numbers are used as weights to aggregate research performance; student numbers are used to weight teaching performance. The performance attributes chosen are those that contribute to international academic standing. We find that the rankings of the Australian National University (ANU) and some technologically oriented universities are raised when scope is allowed for, but elsewhere the changes are minor.

1. I am greatly indebted to Emayenesh Seyoum and Carol Smith for assistance with preparation of the data sets and to the Department of Education, Science and Training and the Graduate Careers Council of Australia for the provision of data.

Notes

1. I am greatly indebted to Emayenesh Seyoum and Carol Smith for assistance with preparation of the data sets and to the Department of Education, Science and Training and the Graduate Careers Council of Australia for the provision of data.

2. Data from Essential Science Indicators relate to the period January 1997 to April 2007. University Statistical Indicators cover the period 1996–2005. In all the USI data the citations relate to two five‐year periods which are then added. I am indebted to Thomson Scientific for permission to use the USI data.

3. The Humanities data combine publications in the USI categories, Classical Studies, History, Philosophy, and Literature.

4. The data are from the Course Experience Questionnaire administered by all universities through the Graduate Careers Council of Australia.

5. For three universities (CDU, CQU, and ECU) TES data were not available for 2005 so 2004 data are used.

6. The payment to the Australian National University for the (research) Institutes is excluded. The mix of disciplines and level of courses (including masters by coursework) makes it difficult to establish appropriate discipline weights, so only aggregate student numbers are used.

7. Publications in quality journals are implicitly given a double weight as they enter in both ISI and DEST publications.

8. The largest fall occurs for Curtin owing to the low staff weight given to its good performance in Education. The result may be due to inappropriate classification of staff in the data.