Underestimating or overestimating the distribution inequality of research funding? The influence of funding sources and subdivision
Research funding is a significant support for the development of scientific research. The inequality of research funding is an intrinsic feature of science, and policy makers have realized the over-concentration of funding allocation. Previous studies have tried to use the Gini coefficient to measure this inequality; however, the phenomena of multiple funding sources and funding subdivision have not been deeply discussed and empirically studied due to limitations on data availability. This paper provides a more accurate analysis of the distribution inequality of research funding, and it considers all of the funding sources in the funding system and the subdivision of funding to junior researchers within research teams. We aim to determine the influence of these two aspects of the Gini results at the individual level. A dataset with 68,697 project records and 80,380 subproject records from the Chinese Academy of Sciences during the period from 2011 to 2015 is collected to validate the problem. The empirical results show that (1) the Gini coefficient for a single funding source is biased and may be overestimated or underestimated, and the most common data source, which is the National Natural Science Foundation of China (NSFC), causes the Gini coefficient to be underestimated; and (2) considering the subdivision of research funding lowers the inequality of research funding, with a smaller Gini coefficient, although the decrease is moderate.