The God has created a man in order that he creates that the God fails to do

Monday, 18 July 2011

Does Impact Factor show anything?

The Impact Factor 2010 of scientific journals has been recently published. At present, IF is considered as an important characteristic of scientific journals. But does it show anything?

Recall that IF2010 of a journal “A” equals N/M where N is the number of citations in 2010 of articles published in “A” in 2008 – 2009 and M is the number of articles published in “A” in 2008 - 2009. For instance, if N=100 and M=100, then IF=1.000.

I am a mathematical physicist. Therefore, let us look at the Table of IF2010 of journals in mathematical physics. The most authoritative journal in this area is Communications in Mathematical Physics. However, it occupies the 12th position with IF=2.000 in comparison with the highest IF=3.144.

It is readily observed that thin journals of about 50 articles in a year a priory have advantages over the thick ones. At the same time, thick journals can widen their scope that gives an advantage over the specialized one. For instance, this is a recent practice of Journal in Mathematical Physics (IF=1.291) and Journal of Physics A (IF=1.641).

IF of a journal in mathematical physics is higher than it is closer to applications and theoretical physics. In particular, the above mentioned Journal of Physics A has IF=1.641, whereas the theoretical journal Classical and Quantum Gravity has IF=3.098.

A problem is that only one well-cited article can essentially increase IF of a journal during two years. For instance, in the example above, let a journal “A” in 2007 published an article quoted 101 times every year. Then its IF in 2008 and 2009 becomes equal to 2.000 in comparison with its usual value 1.000

This is the case of our International Journal of Geometric Methods in Modern Physics. One article on gravitation theory published in our Journal in 2007 and cited 491 times made its IF2008=1.464 and IF2009=1.612 in contrast with 0.662 in 2007 and 0.757 in 2010.

No comments:

Post a Comment