An easy way to boost a paper's citations: Cite and you shall be cited(nature.com)
nature.com
An easy way to boost a paper's citations: Cite and you shall be cited
http://www.nature.com/news/2010/100813/full/news.2010.406.html
3 comments
Totally agree. Correlation does not imply causation.
Frankly, I expected more citations in her "References" section.
It's taken centuries for academics to figure out what bloggers discovered in about five years?
I thought "Oh, so it's like Twitter."
I hate when statistical analyses that turn up correlations are summarized by stories implying a causal link: in this case, “reciprocal altruism”, and even worse, “a single extra reference in an article now producing, on average, a whole additional citation for the referencing paper” (my emphasis), which doesn’t seem at all proven to me. Everything the researcher himself says is of course either completely noncommittal (“If you want to get more cited, the answer could be to cite more people”) or else a platitude (“Scientists are subject to social forces as much as anyone in any other profession” .... really now??!).
To discover a causal link here would require artificially and intentionally (but without telling anyone) adjusting the number of citations in some random set of papers to be published (probably by including or leaving out minor or tangential citations), and then seeing how one group compares to the other in the future.
My speculation about reasons (other than learned altruism) the amount of correlation between references and citations might have changed over time: (1) newer papers are in general treading in more developed fields, in which the top researchers must be familiar with a broader set of previous papers, (2) dissemination of papers is easier than it ever was before and so for example when reading a paper it is trivial to track down all of the papers it references and then cite those directly, (3) expectations about the proper number of citations in a paper have changed over time and modern journal reviewers especially in respected journals are stricter about suggesting that authors cite every previously known idea, (4) relatedly, academia has become increasingly bureaucratically output-driven and so individual researchers are writing more (presumably smaller-individual-impact) papers, (5) if number of citations is increasing over time, and in general most works are cited most often in the few years after their publication, the whole structure operates somewhat like a ponzi scheme, where each generation of papers provides more citations to the immediately previous one than that generation provided to the one before it.