Search toggle

Ranking lists

Rankings are intended to simplify and clarify a complex reality by measuring a number of indicators, often resulting in aggregate values that are meant to reflect a certain aspect of the quality of the universities and their education or research.

Contact

Stefan Carlstein
stefan.carlstein@ju.se
036-10 10 15

Bibliometric indicators are of interest in the present context as they sometimes are included as one indicator of the end result of research.
University rankings have increased strongly during the latest years and several different ranking systems, with both national and international focus, exist today. It is often motivated by the fact that it makes prospective students aware of differences in quality between universities, and thus works as help in deciding which of them to study at. Also, the results of the rankings are considered to stimulate the competition between the universities and constitute a driving force regarding the quality development of higher education.

A stringent definition of ranking systems is hard to make since the differences between them are quite large. However, below are some of the most common aspects that characterize the systems.

  • They are constituted by a number of indicators. It might involve combinations of different input indicators (such as the number of teachers and revenues) and output indicators (such as student throughput or the volume of scientific publishing).
  • The indicators intend to measure quality. The universities found at the top of the ranking lists are regarded, explicitly or implicitly, as being better than the ones at the bottom. The indicators are therefore used to reflect quality. However, no generally applicable definition exists of the quality of higher education and of the research that is carried through. The choice of indicators therefore varies strongly and different interpretations may exist regarding what a given indicator actually measures.
  • The ranking systems describe different types of units. This can apply to the universities as a whole, different units of a university, specific universities or education programs. The most well-known types of ranking systems are those that operate at university level.
  • The indicators are usually weighted together into an aggregate result. The weighting of the indicators is based on the system developer's idea of what is most important and thus should carry the most weight in the total result. Yet sometimes each indicator is presented seperately with the purpose of creating a multidimensional ranking.
  • Ranking implies an order of precedence of the units involved. The units are placed in relation to each other and thus compete for the top place. Consequently, it is never a question of results related to absolute criteria or a quality standard decided upon beforehand.

However, one should be aware of the fact that there exists an extensive critique of ranking lists that should be relevant for the choice of any of the examples above. The most common objection is that the choice of indicators and thus the definition of quality that a given ranking system uses as a basis is not valid, that they are chosen on the basis of easily accessible data rather than on well-founded theoretical arguments. The different weighting methods used for the indicators in question are often regarded as arbitrary and they are criticised for being non-robust in general which might result in unreasonably large changes for a university between two years. Furthermore, the differences between the universities do not necessarily reflect the difference with any statistical significance.

Content updated 2017-07-27






We use cookies on JU.se. By continuing to use this site you accept the use of cookies. More information