Sunday, 19 August 2012

citation patterns

An academic's contribution to the community is made through publications.  And so we obsessively keep good records of what we have published, and neurotically check out if anyone has noticed.

To this end, I’ve been playing around with Google Scholar, tidying up my auto-generated publications list there. Although I still maintain my own publications page, this separate one is a useful resource, mainly because it includes citation counts and links.

Citation links are good for more than just egoboo and feeding neuroses, honestly! Citations indicate other authors who find my work relevant to them, so it's very likely that their work is interesting and relevant to my own research. It makes sense for me to check them out, and then also check out the other papers they reference, and also their citations, to see if there’s something interesting and relevant to me. Although it is a bit easy to get sucked into a never-ending death-spiral of link following: there’s just too much interesting stuff.

That's the beauty of Google Scholar.  it doesn't just have a list of my papers (a task I can do myself with relatively little effort); it also auto-generates links to citing papers (a task that would be ridiculously time-consuming to do myself).

More than that, it also provides neat little graphs of citations per year. These are quite interesting for publications that are several years old, where patterns have had time to emerge.

The typical pattern shows a peak a few years after publication, then a tail off into obscurity.

Slightly more gratifying is the pattern of a peak, followed by a drop, but then carrying on bumping along at about the same rate.

But the weirdest citation pattern must be to a paper written during my PhD, published nearly 30 years ago. It caused barely a ripple when published (I comfort myself that there was no Web in those days, and hence restricted access to it). But then there was a second (modest) peak of citations after about 15 years, followed by a drop-off. And then there was a third peak, starting more than two decades after publication.

Clearly, a paper ahead of its time!

No comments:

Post a Comment