Mike the Mad Biologist

In scientific publishing, one of the important things is what is known as the "impact factor" which is the the average number of citations a journal receives over a 2 year period. The impact factor is often used by librarians and researchers to determine which journals to purchase and where to publish. There are some problems with the impact factor, however.

The basic problem is that the impact factor does not consider the importance of the citing article. In other words, the cited paper is far more likely to be looked up by a reader if it is cited in five high profile journal articles (or five times in one high profile journal, etc.) than if it is cited twenty times in journals that are read infrequently. Yet, according to the impact factor rating, the latter case is more important.

At the University of Washington, Ben Althouse, Carl Bergstrom, and Jevin West have developed a new ranking system which they call eigenFACTOR. This system takes into account how widely read the citing journal is. I've gone to the website and compared the impact factor rankings to the eigenFACTOR rankings for the disciplines I know very well (evolution, ecology, microbiology, and infectious disease). eigenFACTOR seems to reflect how I, and I think my colleagues, rank journals far better than the impact factor.

Go over to eigenFACTOR and see what you think.

I would add that it does some other interesting things:

1) Theses, software, and books are including in the eigenFACTOR rankings.

2) It actually calculates, based on publisher prices, the 'value' of publishing in a given journal.

3) The phylogenetics sofware PAUP* is twelfth ranked evolutionary biology 'journal', which shouldn't be surprising. Nonetheless, it has to be embarrassing to have one's journal 'beaten' by a software program.

Again, it's worth the looksee.

Update: I got my statistics in my bioinformatics--it's eigenFACTOR, not vector...