In late 2005, Jorge E. Hirsch, Professor of Physics at the University of California, San Diego, published an article describing the h-index, which, he called “a useful index to characterize the scientific output of a researcher.” (See J.E. Hirsch, “An index to quantify an individual’s scientific research output,” PNAS, 102(46): 16569-72, 15 November 2005.
This paper (which can be viewed at http://www.pnas.org/cgi/content/abstract/102/46/16569) has been cited 133 times as of June 2008. A survey of the current edition of Thomson Reuters’ Essential Science Indicators database reveals that the h-index is the hottest topic in information science today. Hirsch himself recently published another paper on this subject: “Does the h-index have predictive power?,” (PNAS, 104(49): 19193-8, 26 November 2007).
In his paper, Hirsch describes the h-index as follows: “A scientist has an index h if h of his or her Np papers have at least h citations each and the other (Np – h) papers have less than or equal to h citations each.”
Np is the number of papers published over n years. He found that among physicists he surveyed, Edward Witten, a mathematical physicist at the Institute for Advanced Study in Princeton, New Jersey, and a pioneer in the field of string theory and related areas, had the highest h-index: 110. “That is,” Hirsch explained, “Witten has written 110 papers with at least 110 citations each.”
Hirsch argued that “h is preferable to other single-number criteria commonly used to evaluate scientific output of a researcher,” and he listed “total papers,” “total citations,” “citation per paper,” “number of significant papers” (defined as the number of papers with more than a certain number of citations), and “number of citations to each of a researcher’s q most-cited papers” (for example, q = 5).
Hirsch himself, however, cautioned in a news release issued by his university, that the h-index “should only be used as one measure, not as the primary basis for evaluating people for awards or promotion.” He added, “You surely wouldn’t want to say that in order to get tenure or to get into the National Academy of Sciences you need to have an h-index of such and such.” (See: http://ucsdnews.ucsd.edu/newsrel/science/MCH.asp)
Researchers in the field of infometrics and scientometrics have proposed several variations on the h-index, which, they claim, improves this measure in various ways. The h-index has also been applied to groups of researchers, institutions, nations, and journals.
But is the h-index really superior to other measures, such as total citations?
Henry Small, Chief Scientist with the Scientific business of Thomson Reuters, points out that the h-index “is highly correlated with total citations. In other words, you get almost the same ranking.”
“The convenient feature of h is that if you’re looking at a common name that represents multiple people, then all you need to do is verify that the top n papers are by the same J. Smith, when total cites are not known,” said Small. “The reason h works has to do with the universality of the power law distribution of citations.”
“Of course, one needs to add the usual caution that the h-index, as with other commonly used citation metrics, is highly sensitive to variations in time period and field of science.”