Comments on: Scientific Impact Measures Compared

By: Johan Bollen

Johan Bollen — Mon, 29 Jun 2009 18:51:01 +0000

This paper is now officially published by PLoS ONE:

Bollen J, Van de Sompel H, Hagberg A, Chute R, 2009 A Principal Component Analysis of 39 Scientific Impact Measures. PLoS ONE 4(6): e6022. doi:10.1371/journal.pone.0006022

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0006022

By: Now On the Horizon: Start-up and Apps That Can Change Your World « The Scholarly Kitchen

Fri, 29 May 2009 16:00:34 +0000

[...] how this could be gamed, as Phil Davis showed in a post last year). Phil has also blogged about the problems with usage-based counting overall. This part of Mendeley’s mission seems quixotic to me, and unnecessary. They should focus a [...]

By: Usage Map of Science « The Scholarly Kitchen

Usage Map of Science « The Scholarly Kitchen — Mon, 16 Mar 2009 10:44:50 +0000

[...] Bollen is the principal investigator of the MESUR group, which focuses on usage-based research. Last month, we reported on this group’s work comparing scientific impact measures. [...]

By: Comparing 39 impact measurements | Educationload.com

Comparing 39 impact measurements | Educationload.com — Sat, 21 Feb 2009 15:06:26 +0000

[...] of 39 scientific impact measures, a preprint deposited in arXiv February 12, 2009. (Thanks to Philip Davis.) Abstract: The impact of scientific publications has traditionally been expressed in [...]

By: What the heck is an Impact Factor? « Submitting Your Scientific Manuscript

What the heck is an Impact Factor? « Submitting Your Scientific Manuscript — Tue, 17 Feb 2009 20:57:44 +0000

[...] with a lower IF. However, like most things in life, what sounds good on paper is actually quite complicated. There are piles of studies and commentaries on why the IF is not a reliable metric (here is just [...]

By: Johan Bollen

Johan Bollen — Tue, 17 Feb 2009 15:35:57 +0000

Hi Phil, great summary of our paper. I like your “blind men” analogy, but much depends on the size of the beast and the number of blind men. 200 blind men describing a dog, still a complicated beast, would do much better. Also, nothing is to stop each blind man from feeling various parts of the beast, or in fact the whole animal, and then comparing notes.

The latter may in fact be the correct analogy because each of our metrics is calculated on the entire citation/usage data set, the resulting rankings are compared over all +7000 journals, and our loadings indicate about 85% of all variation is covered by the first 2 components.