Friday, January 11, 2013

ANU is Number One in the World...

On mean kilometres between collaborators on papers! On average our collaborators are 4,522km from us:

This is just one the rankings you can get from the Leiden Ranking of universities. They aim to be more transparent than other well-known rankings like the Shanghai Jiaotong or QS rankings. Another very interesting feature are confidence intervals obtained by boot-strapping.

ANU does not rank very well by more conventional measures of impact. The authors preferred ranking is by the PP top 10% indicator. This is the percentage of a university's publications that are in the top 10% most cited publications. It's not source normalized, however, so universities with strong biomedical research efforts will rank higher. ANU is ranked only 114th with 12.9%. But note that even the top school - MIT - only scores 25.2%. So there is not much variation across most of the 500 research universities.

Another ranking is by source normalized impact factor. This should take into account the differences between disciplines. ANU ranks 118th with an impact of 1.21. MIT has an impact of 2.17. So there is again little variation. Much less variation than among journals which have less variation than individual researchers, which have less variation than articles. I'm actually surprised how little variation there is across universities. Of course, all these are research oriented universities, but still.

Something missing from these ranking is the average number of articles per researcher. It's not possible to work this out from the Web Science in any accurate way. Based on the RePEc rankings Australian economists publish above average numbers of papers given their RePEc rank, but these are cited less. I suspect because of the distance of Australia from elsewhere, despite our high levels of international collaboration.

No comments:

Post a Comment