A proposal for a quantitative indicator of original research output

Abstract: The use of quantitative indicators of scientific productivity seems now quite widespread for assessing researchers and research institutions. There is a general perception, however, that these indicators are not necessarily representative of the originality of the research carried out, being primarily indicative of a more or less prolific scientific activity and of the size of the targeted scientific subcommunity. We first discuss some of the drawbacks of the broadly adopted $h$-index and of the fact that it represents, in an average sense, an indicator derivable from the total number of citations. Then we propose an indicator which, although not immune from biases, seems more in line with the general expectations for quantifying what is typically considered original work. Qualitative arguments on how different indicators may shape the future of science are finally discussed.