The most-cited research of all time
November 13, 2014 6:29 AM   Subscribe

The top 100 papers: Nature explores the most-cited research of all time.

Citations, in which one paper refers to earlier works, are the standard means by which authors acknowledge the source of their methods, ideas and findings, and are often used as a rough measure of a paper’s importance. Fifty years ago, Eugene Garfield published the Science Citation Index (SCI), the first systematic effort to track citations in the scientific literature. To mark the anniversary, Nature asked Thomson Reuters, which now owns the SCI, to list the 100 most highly cited papers of all time. (See the full list at Web of Science Top 100.xls or the interactive graphic, below.) The search covered all of Thomson Reuter’s Web of Science, an online version of the SCI that also includes databases covering the social sciences, arts and humanities, conference proceedings and some books. It lists papers published from 1900 to the present day.

The Web of Science is not the only index of citations available. Google Scholar has also generated a list of the ‘most-cited’ articles of all time for Nature (Google Scholar Top 100.xls). Two-thirds of the entries are books, which Thomson Reuters did not include. “Folks have focused on journals, but there is this other world of books out there,” says Anurag Acharya, a software engineer who leads the Google Scholar team in Mountain View, California. At number 4, the most-cited book is the manual Molecular Cloning, a mainstay of molecular-biology laboratories. But the list shows that research articles can be just as influential as books, notes Acharya. And at the top of both Google’s and Thomson Reuters’ rankings are the same three research articles — albeit in different order.

A separate Google Scholar top 100 showing only the top-cited research articles (Google Scholar Top 100 articles only.xls) throws up many similar papers to the Web of Science ranking. Noticeably, however, just over one-third of the list is different, with economics and psychology articles making considerable inroads, perhaps because they gain more citations from books than do other fields.
posted by mlis (12 comments total) 16 users marked this as a favorite
For Economics alone, the Research Papers in Economics database maintained by the St. Louis Fed (nowadays) has more rankings thank you can shake a stick at, depending on how you want to weight citations.

Glancing over it, a lot of the papers in economics are econometric techniques (Arellano and Bond, the first one on the list, for example, and Heckman's paper near the top) and many of the others are papers that contained a nice modelling trick that gets used a lot later (Dixit and Stiglitz 1977, Calvo 1983 contain features of the workhorse models of imperfect competition and sticky prices, and probably suffer from being so widely known that people don't really feel a need to cite them anymore - everyone knows what the Dixit-Stiglitz aggregator is.). But I'm in macro, so that's my bias.
posted by dismas at 6:41 AM on November 13, 2014

Who ever writes a paper using all 100 as references should automatically get a Nobel Prize.
posted by Renoroc at 6:45 AM on November 13, 2014 [3 favorites]

Before reading: My guess is it's mostly methods and methodologies.

After reading: Yep.
posted by yeolcoatl at 6:50 AM on November 13, 2014 [3 favorites]

And specifically, the top 6 papers and 14 of the top 20 introduce a new tool or technique in molecular biology.
posted by mbd1mbd1 at 6:52 AM on November 13, 2014

I'm of the opinion that method papers are incredibly useful things to have in the literature. They may be an easy way to boost citations, but, frankly, a new method in a procedural lab discipline is absolute gold.

Every postdoc I get, I try to get them to do one method paper. Too many people, imo, stick them as an afterthought in data papers, where they're a poorly described abbreviation, sometimes not even fully described in the supplemental. It limits reproducibility and isn't great science. Publish your methods!
posted by bonehead at 7:10 AM on November 13, 2014 [8 favorites]

I can't see the phrase "of all time" without thinking of Kayne.
Yo Lowry, I know your protein determination assay is cited like mad, but Ijima's paper on carbon nanotubes was the most-cited paper of all time!
posted by exogenous at 7:46 AM on November 13, 2014 [2 favorites]

Publish your methods..... posted by bonehead

See the amusingly abbreviated NMETH for details!
posted by lalochezia at 8:23 AM on November 13, 2014

Who ever writes a paper using all 100 as references should automatically get a Nobel Prize.

Or their Mad Scientist merit badge, at the very least.
posted by Panjandrum at 8:33 AM on November 13, 2014 [1 favorite]

Methods papers are vital. I've seen this over and over in my qualitative research. Too often people use shorthand like "I applied discourse analysis" and then you see what they did based on their short methods paragraph, and it's certainly not discourse analysis in the classic sense, but there's not enough detail to really say what it is they actually did and if it even has a name or precedent. Probably what they did was "exploratory qualitative research with open coding" which is not the same thing - that's like saying "this rectangle is a square" when it is obviously a rectangle that is not a square. Cherry-picking parts of methods and then claiming to follow that method is so, so common and so, so unhelpful in qualitative work. You can't really determine the validity or reliability of a study if the methods aren't clear, and you can't evaluate the methods unless there are people who talk about the method and write about the method and apply said method consistently.

I'm in the process of writing a bulky methods paper right now and it's incredibly challenging. I hope that others read it and get something from it - not so that I am cited (I could not care less about that), but so the method is applied consistently and properly throughout my discipline and other disciplines. So few people do what I do the right way, but many people say they are using the method for their work. That waters down the field, muddies the waters, and makes the research conducted using that method seem much weaker. And it's frustrating to sit down and meet a kindred soul who says they use your method and then when they tell you what they did you have to either shut up or somehow break the news to them that, no, they didn't really do what they say or think they did at all, because they missed some huge steps. It's like throwing out half of the items on a valid survey instrument and then claiming it's still valid. It's not!
posted by sockermom at 1:14 PM on November 13, 2014 [1 favorite]

Methods papers are very, very important, yes, but I would still like to see the top 100 data papers.
posted by gingerest at 3:49 AM on November 14, 2014

Didn't expect it coming from Nature, but we have a new Dumb Journalism Unit: the Mount Kilimanjaro. Apparently if you scale 58 million papers to Mount Kilimanjaro (what the hell does that even mean?) the top 100 would represent only the top 1cm of the mountain. What advantage this has over just saying "only 100 out of 58 million papers are in the top 100 most cited papers" (which is a tautology, of course) I do not know.
posted by Anticipation Of A New Lover's Arrival, The at 10:03 AM on November 14, 2014

Also, it's no surprise to me that most of the most-cited papers are either methods papers or announcements of new software packages. Want people to reference your work? Come up with a tool or a protocol that makes people's lives easier or enables them to get at questions that they couldn't before. Ideally it should be cheap, foolproof, broadly relevant, and let people do stuff that they've always wanted to do but couldn't. Sanger sequencing and PCR are two of the most obvious here in Biology—without them we wouldn't have had the genetics revolution—and lo and behold there they are!
posted by Anticipation Of A New Lover's Arrival, The at 10:07 AM on November 14, 2014

« Older "I still get hate mail from active and retired...   |   Medical Legacy of the War 1914-1918 Newer »

This thread has been archived and is closed to new comments