Abstract
Here we suggest a new index to estimate the scientific impact of an individual researcher, namely the S-index. This index has been designed to emphasize highly cited, truly important works and to be minimally affected by poorly cited ones, without setting any arbitrary threshold. The first property makes it advantageous over the h-index, which does not discriminate between highly and moderately cited articles, while the second property - over the total number of citations, preventing the possibility of overclocking an index by publishing many trivial articles. Contrary to the h-index, which has an upper limit of the total number of publications regardless of their citation numbers, the S-index is not limited by the publication count. This allows scientists having few but very influential works to receive appropriate and respectable index values, which is impossible with the h-index. Moreover, only 10 most cited publications of an individual are typically required to calculate an S-index to 99% accuracy. Collectively, the S-index is principally different from the existing scientometric indicators and should facilitate better recognition of prominent researchers.
