I was wrong about impact factors

Previously, I stated  that "The only value I see in Impact Factors is helping librarians compile draft lists of journals to cancel subscriptions to in order to save money."

Now I don't even think Impact factors are good for that!

This was brought home last week when the UQ library announced that because of the declining Australian dollar that it has to save A$1M and so would be cancelling some journal subscriptions. A proposed "cull" list has been circulated. The exact selection criteria used are not clearly defined but impact factor is stated as one.
Guess what? More Mathematics journals are slated for cancellation than in any other field! 
This is hardly surprising because the average number of citations per article in Mathematics is one third of that in Physics and Chemistry. Thus, IF's for maths journals will be typically smaller by a factor of three.

Also on the cancellation list is the American Journal of Physics and The Physics Teacher. It turns out the former has an impact factor less than one. (I could not find IF for the latter).
This is hardly surprising because Am. J. Phys. largely comprises pedagogical articles that faculty can refer an advanced undergraduate or a beginning graduate student to. [They are also good for faculty who want to learn new things or obtain a deeper understanding of basics.] Such articles are extremely valuable, particularly for an institution that aims to do a good job of engaging undergraduates in research and improving pedagogy. However, they are unlikely to be cited much because they don't contain anything "new" and are not review articles.

Another issue that this journal cancellation exercise highlights is the pernicious practice of "bundling" that Elsevier and some other commercial publishers use. It is all or nothing. Institutions are forced to subscribe to a whole bunch of crap journals in order to get some decent ones.