Research impact evaluation based on effective authorship contribution sensitivity: h-leadership index

The evaluation of a researcher's performance has traditionally relied on various bibliometric measures, with the h-index being one of the most prominent. However, the h-index only accounts for the number of citations received in a publication and does not account for other factors such as the number of authors or their specific contributions in collaborative works. Therefore, the h-index has been placed on scrutiny as it has motivated academic integrity issues where non-contributing authors get authorship merely for raising their h-index. In this study, we comprehensively evaluate existing metrics in their ability to account for authorship contribution by their position and introduce a novel variant of the h-index, known as the h-leadership index. The h-leadership index aims to advance the fair evaluation of academic contributions in multi-authored publications by giving importance to authorship position beyond the first and last authors, motivated by Stanford's ranking of the top 2 \% of world scientists. We assign weighted citations based on a modified complementary unit Gaussian curve, ensuring that the contributions of middle authors are appropriately recognised. We apply the h-leadership index to analyse the top 50 researchers across the Group of 8 (Go8) universities in Australia, demonstrating its potential to provide a more balanced assessment of research performance. We provide open-source software for extending the work further.