Ranking Scientists

Abstract

Purpose: I provide a framework to construct rankings of scientists based on journal articles and their citations. Design/Methodology/Approach: I assume a model in which the quantity and the impact of publications are economic goods and scientific committees derive utility from them. The committees’s utility is therefore defined on the set of ordered pairs of the type (k, ck ) interpreted as “the kth most important publication of a given author has ck cites.” Within this framework, I derive the performance measure induced by utility maximization. Findings: I prove that when quantity and impact are perfect substitutes, the induced performance index is the w-index in Woeginger (2008a) [“An axiomatic characterization of the Hirsch-index,” Mathematical Social Sciences, 56, 224—232.] In the case where quantity and impact are perfect complementaries, the induced performance index is the well known h-index in Hirsch (2005) [ “An index to quantify an individual scientific research output,” Proceedings of the National Academy of Sciences, 102 (46), 16569—16572.] Finally, when preferences are of Cobb-Douglas type, i.e. the trade off between quantity and impact equals the ratio between papers and cites, the induced index is Komulski’s 2007) maxprod index [“MAXPROD—A new index for assesment of the scientific output of an individual, and a comparison with the h-index,” International Journal of Scientometrics, Informetrics, and Bibliometrics,, 11 (1).] Research limitations: The analysis of this model does not include some widely extended measures, as the criteria of average citations per paper. Originality/Value: This model allows for a re-examination in terms of academic preferences of some scientific impact measures. Conversely, the model can help ranking designers to fit the needs of the institution using the ranking by first calibrating a utility function and then find the ranking induced by this utility.