The harmonic h-index: an impromevement of the Hirsch h-index and the Egghe g-index

TL;DR

This paper introduces the harmonic har-index, a new metric for scientific output that is easy to compute, correlates well with existing indices, and aims to combine their advantages while reducing disadvantages.

Contribution

The paper defines the harmonic har-index, proves its bounds relative to h and g indices, and demonstrates its high correlation with the hg-index, offering a simpler alternative.

Findings

har-index is highly correlated with hg-index

har-index is easier to compute than g and hg indices

har-index maintains advantages of h and g indices

Abstract

In order to characterize the scientific output of scientists, in this paper we define the harmonic har-index whose values are positive integers. It is proved that $h \leq har \leq g$, where h is the Hirsch index and g is the Egghe index. Despite the fact that the har-index is defined in a completely different way (including sum of reciprocals of citations of a researcher), based on our computatioanl results, we get the surprising fact that this index is highly correlated with the hg-index ($hg=\sqrt{hg}$ ) introduced by Alonso et al. (2010). Accordingly, we believe that the har-index keep the advantages of both measures as well as to minimize their disadvantages. In addition, it is much easier to calculate the values of har-index than those of g-index and hg-index.