We’ve had some discussion recently about the use of bibliometrics and impact factors as a way of evaluating research quality. Very recently, my collaborators chose to submit our joint paper to one journal rather than another because the chosen journal had a higher impact factor.

Today I found this interesting page, which gives the impact factors of physics journals:

http://www.tcm.phy.cam.ac.uk/~tmf20/physics_journals_impact_factor.shtml

You can see some interesting disciplinary differences even from this list: particle physics journals tend to have higher impact factors than condensed matter journals, for example. It is also interesting that overall there are very few physics journals with what one might call high impact factors: Phys. Rev. Lett., Nature Physics, and some review journals. Most of the rest (except for the particle physics journals) tend to have impact factors below 3. In contrast, many of the “normal” biochemistry journals have impact factors around 5.

This list doesn’t include a lot of the fluid mechanics journals I tend to publish in these days but they all have impact factors around 2 to 2.6.

What does this all mean? Your guess is as good as mine. But here is one interpretation (not mine!): Physics doesn’t have much impact.

Of course it is easy to raise the impact factor of the disicpline as a whole. All physicists simply have to cite twice as many papers as we normally would when we write our own papers. Of course it would also help if we all wrote twice as many papers as well, all of which cited lots of other papers. The result would be a huge number of useless papers with meaningless references, but Physics’s impact factor would go up. This should give you some idea of my view of the value of this statistic as a measure of the quality of research.