31 January 2017

Wake up calls for scientific papers

I once called scientific articles “love letters to the future.” I was making the point that just because a paper doesn’t get cited soon after its publication does not mean that it is useless.

A new article by Ho and Hartley bears out that description. It focuses on three papers that were largely uncited for many years, the, for some reason, it was noticed and became widely cited.

The authors don’t use my romantic term. There is another established term for these papers, also with romantic overtones: “sleeping beauties.” They pick out three, from the field of psychology, which hasn’t been studied before. This one has the most dramatic citation shift, really only picking up steam when it reached retirement age (65 years after publication!)

While it is encouraging to know that these reversals of fortune can happen, Ho and Hartley show that they are extremely rare. They found only three cases out of over 300,000 psychology papers (0.001%).

The duo don’t identify who re-popularized these papers. It’s a little frustrating, because the paper is pretty short.

Hat tip to Remi Gau and Neuroskeptic.


Ho Y-S, Hartley J. 2017. Sleeping beauties in psychology. Scientometrics 110(1): 301-305. http://dx.doi.org/10.1007/s11192-016-2174-0

Related posts

Better a deluge than a drought

Picture from here.


Marklar said...

Thanks for posting this. It's a wonderful antidote to the oft-experienced neurosis of imposter syndrome that haunts so many of us, especially us middling ones with modest citation indices.

Dennis Burke said...

To your last point about what caused these papers to take off, I was playing around with Google scholars functions that allow one to search around by keyword and within date ranges for texts that cite a given article. I think this can give some insight into what exactly popularized these articles.

For instance the 1935 Stroop article's 6th or 7th citation comes from a book in 1950 by HJ Eysenck called "Dimensions of Personality," which itself has been cited >2200 times. This is the first widely cited work that cites this article. Though 15 years after Stroop's pub barely registers a blip on the graph of citation counts, 30 years is where it really starts to take off. Right about that time in 1962 is another book called "Psychological differentiation: Studies of development" by Witkin et al. that's been cited >3700 times, which cites the Stroop paper. Other highly cited works begin to cite the Stroop paper later in the 1960s, possibly due to the influence of the Witkin et al book (all speculation).

Without knowing anything about the field or reading these books, my n=1 take away from playing around with Google scholar seems to be: "If you want your paper to take off later in its life, make sure an important and widely read book cites and/or discuss it's content"

This game would be a little more fun if all the works were accessible online to actually skim through. Would like to see how they cite the Stroop work to more fully understand this phenomenon. Would be interesting to see if the 1950 book kept Stroop's paper the ethos long enough for it to be cited by the 1962 book and start to take off, or if they're unrelated.