Statistical Regularities in the Rank-Citation Profile of Scientists
Alexander Michael Petersen
IMT Lucca Institute for Advanced Studies; Boston University
H. Eugene Stanley
Boston University - Center for Polymer Studies
Istituto per le Applicazioni del Calcolo; University of Freiburg
March 14, 2011
Scientific Reports, Vol. 1, p. 181, 2011
Recent “science of science” research shows common regularities in the publication patterns of scientific papers across time and discipline. Here, we analyze the complete publication careers of 300 scientists and find remarkable regularity in the functional form of the rank-citation profile c_i(r) for each scientist i = 1...300. We find that the rank-ordered citation distribution c_i(r) can be approximated by a discrete generalized beta distribution (DGBD) over the entire range of ranks r, which allows for the characterization and comparison of c_i(r) using a common framework. The functional form of the DGBD has two scaling exponents, β_i and γ_i, which determine the scaling behavior of c_i(r) for both small and large rank r. The crossover between two scaling regimes suggests a complex reinforcement or positive-feedback relation between the impact of a scientist’s most famous papers and the impact of his/her other papers. Moreover, since two scientists with equivalent Hirsch h-index values may have different c_i(r) profiles, our results demonstrate the utility of the β_i parameter in conjunction with h, in order to quantify a scientist’s publication impact. We use the analytic properties of the DGBD to derive an exact expression for the crossover value r∗ which determines the set of distinguished papers for a given author.
Number of Pages in PDF File: 29
Keywords: scientific impact, rank distribution, citations, career achievement, Zipf Law
JEL Classification: D39, J24, J44Accepted Paper Series
Date posted: July 21, 2011 ; Last revised: December 5, 2011
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo8 in 0.297 seconds