Although universities’ world rankings are popular, their design and methods still request considerable elaborations. The paper demonstrates some shortcomings in the Academic World Ranking of Universities (ARWU, Shanghai Jiao Tong University) ranking methods. One deficiency is that universities’ scale differences are neglected due to omitting the whole input side. By resampling and reanalyzing the ARWU data, the paper proposes an input-output analysis for measuring universities’ scientific productivity with special emphasis on those universities which meet the productivity threshold (i.e. share of output exceeds share of input) in a certain group of universities. The productivity analysis on Scandinavian universities evaluates multidisciplinary and specialized universities on their own terms; consequently the ranking based on scientific productivity deviates significantly from the ARWU.
[1]
Marguerite Clarke,et al.
Quality Assessment Lessons from Australia and New Zealand
,
2005
.
[2]
Jan Sadlak,et al.
Higher Education Rankings: Evolution, Acceptance, and Dialogue
,
2005
.
[3]
Anthony F. J. van Raan.
Reply to the comments of Liu et al.
,
2005,
Scientometrics.
[4]
Li Liu,et al.
Academic ranking of world universities using scientometrics - A comment to the “Fatal Attraction”
,
2005,
Scientometrics.
[5]
David Jobbins.
Moving to a Global Stage: A Media View.
,
2005
.
[6]
Anthony F. J. van Raan,et al.
Fatal attraction: Conceptual and methodological problems in the ranking of universities by bibliometric methods
,
2005,
Scientometrics.
[7]
Nian Cai Liu,et al.
The Academic Ranking of World Universities.
,
2005
.