Essay: The Methodology behind the Predictions
The Methodology behind the Thomson Reuters Nobel Laureates Predictions since 1989
A Preface to Choosing Thomson Reuters Citation Laureates: The Process and the Results
by David Pendlebury, Citation Analyst,
Science business of Thomson Reuters
Since 1989, the Science business of Thomson Reuters has correctly predicted at least one Nobel Laureate each year with the exception of the years 1993 and 1996. David Pendlebury's predictions, combined with those of Dr Eugene Garfield — Founder and Chairman Emeritus of the Institute for Scientific Information (ISI), now part of Thomson Reuters — have shown that citations in the journal literature reveal influential researchers who have contributed significantly to scientific advances, such as those recognized by the Nobel Prize committees (see Successful Predictions Table).
Two major factors contribute to Thomson Reuters's accuracy in predicting Nobel Prize winners. Firstly, Dr Garfield has for many years studied the correlation between high citation frequency and the receipt of prestigious prizes, especially the Nobel Prize. The primary source for Thomson Reuters's forecasts is the Web of Science®, which provides quick access to the world's leading citation databases with multidisciplinary coverage from over 20,000 of the most influential, high impact journals and conference proceedings worldwide, including Open Access journals, in every area of the natural sciences, social sciences, and arts and humanities. Important scientific discoveries as well as emerging research fronts are always captured. No discovery of scientific importance is overlooked in the Web of Science.
The second reason is Pendlebury's methodology. By selecting the top 0.1% of papers of each scientific field, the scope of analysis can be narrowed to topics and people that are most likely to be nominated by the Nobel selection committee. Citation analysis allows Pendlebury to trace the true pioneer of each field or area of research as well as to assess peer recognition of the discovery.
The value of Web of Science to researchers worldwide lies in it being "selective, yet comprehensive." Web of Science is selective in its journal coverage, to keep searches and results focused and relevant, yet comprehensive enough to cover the most significant scientific research in all disciplines across a long time span.
According to Garfield's "Law of Concentration" which was published in Current Contents #17 in 1971, "a basic list of 500-1,000 journals will account for 80 to 100% of all journal references" in all "special" Bradford distributions. Therefore, Bradford's Law of Scattering1 is generalized according to Garfield's law which suggests "that in opposition to scattering, a basic concentration of journals is the common core or nucleus of all fields. In other words, what Bradford's law postulates for single disciplines, Garfield's law postulates for science as a whole. And it holds no matter how one considers the journals — as a source of citing articles or as a collection of cited articles."
Most important for the Thomson Reuters Nobel predictions, as well as for those searching the literature, is the depth of time captured in the Web of Science, which extends back to 1900 in the sciences. These Back Files provide access to millions upon millions of source items and cited references that include virtually every major scientific advance that has shaped our world — including most of the works that have been honored in the more than 100-year history of the Nobel Prize itself. This coverage enables researchers to conduct deeper and more complete research and to track trends and patterns over time rather than relying on data from a decade or less. For example, the work recognized by the Nobel committee in its Prize for Physics in 2009 dates from the 1960s and 1970s. And the research that earned the Nobel Prize in Physiology or Medicine in 2009 dates to the 1980s.
Science Citation Index Expanded™ — from 1900
Social Sciences Citation Index® — from 1900
Arts & Humanities Citation Index® — from 1975
Current Chemical Reactions® — from 1986, plus archives from 1840-1985 from the French Patent Office (INPI)
Index Chemicus® from 1993
Finally, it is noteworthy that Thomson Reuters is the only organization in the world to use quantitative data in its analysis (largely derived from Web of Science and Century of Science) to develop its annual predictions of the most influential researchers in physics, economics, chemistry and physiology or medicine (Citation Laureates) who go on to receive a Nobel Prize.
1. Testa, J. The Scientific Journal Selection Process (Science business website, Thomson Reuters, 2008).