Scholar performance assessment plays an important role in reward evaluation, funding allocation, and promotion and recruitment decisions. However, raw publication counts and raw citation count-based scholar performance assessment indicators, such as H-index or author citations, have shortcomings; for example, they ignore the impact of different citation patterns under different research topics, leading to authorship credit inflation due to full citation allocation to each author in multi-author publications. This study proposes a new scholar performance assessment indicator called the normalized scholar academic productivity (NSAP) indicator, which overcomes the issues posed by raw citation counts and publication counts-related scholar performance indicators by considering diverse aspects of scholar research achievements. The NSAP indicator considers the research topic, author sequence and author role in the author list, field-normalized journal impact when allocating citation counts to scholars, and published time. The research topic is generated by the co-keyword embedding and semantic relatedness of each keyword in order to make NSAP topic-dependent; the author sequence and role affect authorship credit allocation strategy; and field-normalized journal impact was used to assign different weights on raw publication counts and citation counts. Finally, awardees of the Derek de Solla Price Memorial Medal and the Association for Information Science and Technology's awards were used to evaluate the validity of NSAP for calculating scholar performance assessment. Results reveal outstanding topic-related scholar performance assessment properties compared to raw citation count indicators, such as H-index, author citations, and cited-by counts (i.e., total number of citing authors).
|Journal||Journal of Informetrics|
|Publication status||Published - 2021 Nov|
Bibliographical noteFunding Information:
This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea ( NRF-2020S1A5B1104865 )
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Library and Information Sciences