Choosing Thomson Reuters Citation Laureates: The Process and the Results
By David Pendlebury, Research Services, Thomson Reuters
Updated September 2008
In anticipation of the Nobel Prize announcements for 2008, which will begin on October 6, Thomson Reuters (NYSE: TRI) is once again revealing its own list of Laureates — in this case, Thomson Reuters Citation Laureates.
Thomson Reuters Citation Laureates have been cited so often in the last two or more decades that these scientists typically rank in the top 0.1% in their research areas. Not only do our Citation Laureates have stratospheric citation totals, they also typically write multiple high-impact reports, and do so over many years.
Correlating citations and peer esteem
Numerous studies in the past three decades have shown a strong correlation between citations in the literature and peer esteem, often reflected in professional awards, such as the Nobel Prize. This should cause no surprise. Citations have been likened to repayments of intellectual debts, so persons who have accumulated such credits from their peers are often those whom these peers nominate for prizes and other honors.
It is clear that the Nobel Committees’ selection processes are more complex than simply identifying highly cited or most-cited scientists. Generally, a Committee looks for an area of research to recognize, and then identifies the key persons responsible for the advance, even if the course of selection is determined by dossiers on individual scientists nominated by their peers. As Harriet Zuckerman, the sociologist of science and author of a fundamental study on this subject, Scientific Elite: Nobel Laureates in the United States (New York: Free Press, 1977), has pointed out: “Every year, more scientists are eligible for Nobel Prizes than can win them.” She continues, “This means that there has always been an accumulation of 'uncrowned' laureates who are peers of the prize-winners in every sense except that of having the award” (see page 48).
Thus, in choosing our Citation Laureate “picks” for the Nobel Prize in 2008 (or thereafter) we looked first at citation counts and at number of high-impact papers, but then also at discoveries or themes that might be considered worthy of special recognition by the Nobel Committee. In each of four scientific areas — Physiology or Medicine, Physics, Chemistry, and Economics — we have made three new choices. But as a supplement to this, and in recognition of Dr. Zuckerman's observation, we are also providing the names of those we have ‘tipped’ previously who have yet to win the Nobel Prize. If one or more of our picks should win the Nobel this year — as two of four did last year when Thomson Reuters (then Thomson Scientific) made its choices (Medicine and Physics, see below) — it is more luck than skill. But by focusing on the most-cited scientists, we hope to, as it were, better our luck.
Looking back at previous years' choices: Do Citation Laureates become Nobel Prize winners?
(Editor’s note: Thomson Reuters began formally announcing its Citation Laureates in 2002, but in years prior, and using the same methodology as is currently used, Thomson Reuters executives had informally named Citation Laureates in various outlets — including the pages of The Scientist. This essay reviews the successes in these pre-2003 predictions as well as those that have been made since the formal predictions were announced. For a list of the successful predictions that were made since 1989, visit http://scientific.thomsonreuters.com/nobel/success/.)
In 1989, in the pages of The Scientist (Vol. 3: 14, 2 October 1989), I drew up a list of 20 highly cited researchers who might win the Nobel Prize in Physiology or Medicine, based on the methodology described above. That year, two of the 20 suggested names were chosen for the Nobel: Harold E. Varmus and J. Michael Bishop. Subsequently, four more of the 20 received the Prize in later years: Edwin G. Krebs in 1992, Alfred G. Gilman in 1994, Eric R. Kandel in 2000, and Sydney Brenner in 2002. Of course, while six were correct choices, 14 were wrong--or at least wrong to date.
In 1990, Angela Martello, also writing in The Scientist (Vol. 4: 16, 3 September 1990, and Vol. 4: 16, 17 September 1990), suggested possible Nobel Laureates using the same methodology. She named 12 researchers for the Physics Prize and 10 for the Chemistry Prize. The Physics selections have produced two Nobel Laureates in Physics (David J. Gross and Frank A. Wilczek in 2002), and another of the names selected — Alan J. Heeger — won the 2000 Nobel Prize in Chemistry. So, right person, wrong prize.
Her selections for the Chemistry prize matched four subsequent Nobel winners. In the same year, 1990, Elias J. Corey won the Prize. In 1994, another pick of Martello, George A. Olah, was honored. Then in 1998, John A. Pople won the Nobel. Finally, in 1999, Ahmed H. Zewail, who had been mentioned as a younger Nobelists-to-Be, won the Chemistry Prize.
Again, chemistry: In 1997 Pendlebury circulated by email a list of “ISI's 50 Most Cited Chemists, 1981-June 1997, Ranked by Total Citations.” (see http://pcb4122.univ-lemans.fr/citation.html). This was subsequently posted, along with more extensive rankings, by Professor Armel Le Bail of France. Of these 50, seven had already won the Nobel Prize. Since its release in 1997, five more became Nobel Laureates: Pople in 1998, Zewail in 1999, Ryoji Noyori and K. Barry Sharpless in 2001, and Richard R. Schrock in 2005.
High citation counts have proven a strong predictor of who may win the Nobel Prize in Economics. In 1990, in an essay for Current Contents (no. 11, 12 March 1990, pp. 3-7) entitled “Who Will Win the Nobel Prize in Economics? Here's a Forecast Based on Citation Indicators,” Garfield published a list of the 50 most-cited economists, 1966-1986, which was based on first-author citation data only. (see: http://garfield.library.upenn.edu/essays/v13p083y1990.pdf) This list contained the names of 15 economists who had already been awarded the Nobel Prize. Subsequently, seven more went on to win: Ronald H. Coase in 1991, Gary S. Becker in 1992, Robert E. Lucas in 1995, Amartya Sen in 1998, Joseph E. Stiglitz in 2001, and Clive W.J. Granger in 2003, and Edmund S. Phelps in 2006. (Note: The 51st name — not published — was Robert C. Merton, who won the Nobel Prize in 1997). Thus, 22 names (44%) on the Garfield list have now won the Nobel Prize.
In 2002, I listed three choices for each of the four Nobel Prizes in Physiology or Medicine, Chemistry, Physics, and Economics in the in-cites website of Thomson Reuters Essential Science Indicators database. One of the choices for Economics was correct that year: Daniel Kahneman.
In 2003, I again offered my predictions, which Thomson Reuters (then Thomson Scientific) subsequently publicized and posted to its corporate Web site. Again, two of our nominees for the Nobel Prize in Economics were correct — Clive W.J. Granger of University of California, San Diego, and Robert F. Engle of New York University School of Business.
In addition, the 2003 predictions came quite close to a winner in Chemistry. One of our Thomson Reuters Citation Laureates was Adriaan Bax of the National Institutes of Health in Bethesda, Maryland, for, as our citation read, "revolutionary advances in the use of nuclear magnetic resonance (NMR) to reveal the structure of large proteins in solution." The Prize went instead to John B. Fenn of Virginia Commonwealth University, in Richmond, Virginia, and Koichi Tanaka of the Shimadzu Corporation in Kyoto, Japan (one-quarter each); the other half went to Kurt Wuthrich of the Swiss Federal Institute of Technology in Zurich and of The Scripps Research Institute in La Jolla, California. In claiming to have come close, we note the citation for Wuthrich's portion of the Prize: "for his development of nuclear magnetic resonance spectroscopy for determining the three-dimensional structure of biological macromolecules in solution." So, the citation data led us to the right field but not the right person, at least as the Nobel Committee saw it. In the description of Wuthrich's research mention is made of Bax, who in the 1990s contributed to the advance of Wuthrich's approach. We'll take this as a near miss.
In 2004, Gross and Wilczek (along with H. David Politzer) won the Physics Prize. As noted above, both were featured in Martello’s article of 1990. None of our new picks, however, turned up in the Nobel committee’s selections that year.
In 2005, one of our three new selections for the Chemistry Prize proved correct: Robert H. Grubbs.
In 2006, Edmund S. Phelps won the Economics Prize. He was one of the 50 most-cited economist listed by Garfield in 1990.
2007 was something of an annus mirabilis for us: We accurately predicted two of the four Nobel Prizes — in Physiology or Medicine and in Physics — based on the names offered up in 2006. In Physiology or Medicine, the Prize went to Mario R. Capecchi, Sir Martin Evans, and Oliver Smithies for their contributions to gene targeting (homologous recombination). And in Physics, the Prize went to Albert Fert and Peter Gruenberg for their discovery of the Giant Magnetoresistance Effect.
In summary, based on three articles in The Scientist (in 1989 and 1990), and two lists of the 50 most cited researchers in economics (1990) and chemistry (1997), as well as our very limited choices for each field posted on our website each year since 2002, our successes have been significant: Each year since 1989, with the exception of 1993 and 1996, citation analysis combined with other considerations has led us, time and again, to researchers who later go on to receive a Nobel Prize.