最近，《Nature》杂志发表了中国学者Yuxin Zhao的评论文章，该文指出了中国科研评价体系存在的主要矛盾：尽管中国的科研机构鼓励研究人员遵循国际科研标准，但在评估研究成果时，却采用了国内特定的评价体系。

原文：Zhao Y. Why China needs to review its approach to research evaluation [J]. Nature, 2025, 641（8062）：283-283.

China’s efforts to redefine how scientific merit is measured are the subject of fierce debate. At the heart is a contradiction：Chinese institutions encourage researchers to follow global scientific norms, but they measure performance against locally defined priorities. Many scientists, especially those at early career stages, find it hard to navigate this system with its conflicting incentives.

In March, for example, the National Science Library of the Chinese Academy of Sciences（CAS）updated its journal-ranking system — which complements other standards such asimpact factors and Clarivate’s Journal Citation Reports. Using article citations, its database is re-evaluated each year with a complex algorithm. Journals might move up or down the tiers. However, in 2025, some elevations of domestic titles over prominent international ones caught scholars’ attention. For example, CAS’s Chinese Physics Letters moved to tier 1, whereas the American Chemical Society’s Nano Letters（a global leading title in nanoscience）dropped to tier 2（see go.nature.com/4atq2xx）. Critics attribute such changes to opaque metrics and disciplinary reclassifications — for example, materials science can be classed as engineering or as physics.

Such assessments are well intentioned, but can have unintended consequences. For example, the CAS library states that its system, which is designed to assess journals’ scholarly impact and to inform institutional research-management strategies, should not be applied to evaluations of individual researchers. Yet evidence suggests that scholars in China choose where to publish to align with this ranking. For example, between 2015 and 2019, journals that were downgraded in the CAS list saw a 14.8% decline in articles by authors in China1. Scholars at less-prestigious institutions were more likely to do so to ensure that their career progressed than were those at more-prestigious ones.

Anecdotally, colleagues have told me that work they published in prominent overseas journals received less attention in promotion reviews than did studies published in local ones. Some researchers refer to submissions to certain domestic journals as ‘academic tribute’ — not out of disrespect but because of frustration with the current evaluation frameworks.

If these trends continue, concerns might arise over researchers’ ability to connect globally. That might further dissuade top researchers from coming to China to pursue self-directed innovative work. For example, one 2023 study found that 89% of expatriate top-tier scholars who received offers through China’s Thousand Talents programme decided to continue their careers abroad2. Three measures could help stakeholders to realign incentives with scientific universality while respecting national contexts.

First, establishing international jointly designed evaluation frameworks through organizations such as the International Science Council could create barriers to prevent the politicization or gaming of metrics. Such systems should be secure and transparent. For example, the blockchain-based peer-review trial by the US National Committee for the International Union of Pure and Applied Chemistry embeds in its reviews time-stamped, cryptographically secured records, which can be checked and verified by anyone3.

Second, creating protected intellectual zones — similar to special economic regions but for knowledge production — can allow alternative evaluation models to be stress-tested. Shenzhen’s Qianhai Cooperation Zone, for example, permits full foreign ownership of research and development laboratories in non-restricted sectors and streamlines cross-border fund transfers through its free-trade account system to attract global researchers. In Hangzhou, Westlake University’s experiment with autonomous governance structures, resulting in higher Nature Index contributions per faculty member at this private university compared with state counterparts（seego.nature.com/3ebfw9n）, suggests that such models could coexist with national systems.

Third, reinvigorating peer-led quality assessment requires rebalancing administrative oversight with disciplinary expertise. In Shanghai, Fudan University’s pilot Merit Review System, which evaluates anonymized manuscripts rather than journal brands, offers a promising template. By prioritizing high-impact research and discouraging questionable practices that could lead to retractions, it demonstrates how quality-focused evaluation can align with institutional goals4.

But there are some practical barriers. Some colleagues question whether international frameworks might impose foreign standards. Others note that protected zones could exacerbate regional disparities if they are not carefully designed. These concerns underscore the need for incremental, evidence-based reform that builds on China’s existing strengths while addressing systemic pressures.

No algorithm can engineer Nobel prizes, no bureaucracy can mandate breakthroughs. The path forward demands openness — and humility.

Nature641, 283（2025）

References

1.Sun, Z., Zhang, C., Pang, K. L., Tang, Y. & Li, Y. Scientometrics129, 7035–7054（2024）.

2.Shi, D., Liu, W. & Wang, Y. Science379, 62–65（2023）.

3.Lawlor, B. et al.Pure Appl. Chem.97, 279–330（2025）.

4.Wang, X. & Wang, W. L. Front. Psychol.14, 1271110（2024）.