A critical review of the proposed reforms of the academic performance indicators applied in the assessment of researchers' performance in Hungary

TL;DR

This paper critically reviews proposed reforms to Hungary's academic performance indicators, highlighting issues with current methods, especially the bias towards natural sciences, and suggests more discipline-sensitive assessment strategies.

Contribution

It provides a bibliometric analysis demonstrating the need for discipline-specific co-authorship considerations in research assessment.

Findings

Homogenization of indicators favors natural sciences.

Integer counting disadvantages social sciences.

Discipline-specific co-authorship patterns are crucial for fair assessment.

Abstract

The academic performance indicators of the Doctor of Science title, the highest and most prestigious qualification awarded by the Hungarian Academy of Sciences (HAS), are key in the national assessment system. The types of performance indicators, as well as their minimum values, are incorporated into the application requirements of academic promotions, scientific qualifications, and research scholarships. HAS has proposed a reform of these performance indicators, to align with the current national and global trends. The proposed modifications are generally based on arbitrary decisions and the consensus between academicians, namely, the representatives of the sections of HAS. This paper contains a bibliometric analysis of 25,000 publications produced between 2011 and 2020 by 683 researchers affiliated with HAS's Section of Earth Sciences. The bibliometric data of the publications are processed by integer and fractional counting, respectively. The main goal of the paper is to argue that discipline-specific co-authorship patterns should be accounted for in the assessment procedure. It is also shown that the homogenization of the performance indicators and the rigid use of the integer counting method favour hard natural science disciplines and disadvantage social science disciplines. Finally, I describe some components of an alternative publishing strategy which would be most prudent for researchers, given the proposed assessment criteria.