news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

IntellectualsThe Intellectual

Image source: pexels

Written by Zhang Tianqi

With the rapid development of Chinese science and technology, China's paper output and the proportion of highly cited papers have ranked first.

In June this year, The Economist's cover feature "The Rise of Chinese Science" attracted attention from all parties. The title of an article in the feature bluntly stated that "China has become a scientific superpower." According to the article, in 2003, the number of highly cited papers in the top 1% of citations in the United States was 20 times that of China. The latest data since 2022 shows that China's share of highly cited papers has surpassed that of the European Union and the United States, ranking first.

In fact, China is also growing rapidly in terms of the number of times its papers are cited.

According to the 2023 China Science and Technology Paper Statistics Report by the Chinese Institute of Scientific and Technological Information of the Ministry of Science and Technology, from 2013 to 2023, China's international scientific papers were cited 67.4823 million times, an increase of 18.3% compared with the previous statistical citation number, ranking second in the world, and the United States still ranked first. Among the 22 disciplines, China ranked first in the world in the number of citations of papers in five fields: agricultural science, materials science, chemistry, computer science and engineering technology [1].

From 2007 to 2017, the total number of citations of China’s international scientific papers surpassed that of the United Kingdom and Germany for the first time, ranking second in terms of citations[2]. Seven years later, China’s total number of citations is already about four times that of the United Kingdom and Germany. At the current growth rate, it is only a matter of time before China’s number of citations of international scientific papers becomes the first[2].

However, a report from Japan last year poured cold water on the rapid growth of Chinese citations. According to the Science and Technology Indicators 2023 report by the Institute of Science, Technology and Academic Policy of the Ministry of Education, Culture, Sports, Science and Technology of Japan, Chinese researchers have a strong tendency to cite each other's papers. One of the driving forces behind the rapid growth of China's international scientific and technological paper citations may be this mutual citation between domestic scholars.

Data shows that when analyzing the institutions of researchers who cite international papers by country, Chinese researchers account for the highest proportion of citations from their own institutions, reaching 61%. The United States is close behind, with researchers from Japan accounting for 29% of citations from their own institutions. In comparison, Japan accounts for 19%, while researchers from the United Kingdom, Germany, France, and South Korea account for a lower proportion of citations from their own institutions, at around 10%[3].

This year, a working paper published in NBER also made a similar point: China's growth in scientific literature citations has been overestimated. Among all countries, China has the most obvious home bias.

Is the high number of citations of Chinese papers really due to the mutual citations and group cooperation among domestic scholars?

Do Chinese scientists prefer to work in groups?

The study was conducted by Qiu Shumin, a lecturer at the School of Business of East China University of Science and Technology, Pierre Azoulay of the MIT Sloan School of Management, and Claudia Steinwender of the Department of Economics of the University of Munich. Through Clarivate Analytics' Journal Citation Reports (JCR), the authors selected 461 top journals in the top 10 of each field from 2000 to 2021 to measure the citations of papers published in these journals.

The study found that among all major countries and almost all scientific fields studied, China showed a clear home bias in paper citations. That is, even after taking into account the scale of scientific research paper output in various countries, domestic research papers are more likely to be over-cited by domestic researchers.

According to the authors, Chinese papers account for the largest proportion of citations in China, at 57.2%. A large proportion of citations from the United States also come from within the country, at 37.1%. This distribution is similar to the aforementioned Japanese report. However, China and the United States have a large number of researchers, and given their scale, they are more likely to cite themselves. To correctly measure local bias, we need to measure the deviation of the "citation ratio" of papers from the "production ratio" of papers (the proportion of papers published by a country in the world).

After adjustment, China's home bias is 42.3%, still the most obvious among all countries, while the United States is 15.9%. This phenomenon is not limited to a specific subject area. In 18 of the 20 scientific categories, China shows the strongest home bias. From a temporal perspective, China's paper citations have had obvious home bias since 2000, and the home bias has been increasing over the past 20 years, and by 2018 it had exceeded twice that of the United States.

If the local bias of citations affects the total number of citations, how big is this impact? In terms of the total number of citations of papers, after adjusting for the local bias, the total number of citations of Chinese top journals ranks fourth in the world, with the top three being the United States, the United Kingdom, and Germany.

Qiu Shumin added that if the number of citations of papers is counted over a ten-year period as in the "Statistical Report on Chinese Science and Technology Papers", China's total citations from 2007 to 2017 would still be second in the world, but after adjusting for local bias, the gap with the United States, which ranks first in citations, is even larger, at 23% of the United States.

In terms of research fields, the United States leads in citations in all fields, while China ranks second in chemistry, materials, engineering, computer science, and agricultural science, but has not yet become first. China’s average citations per paper have been hovering between 12th and 15th, far from the top tier [4].

But today, China's scientific research output ranks among the best in the world in terms of both quantity and quality of papers. However, China has the most obvious local bias, which means that compared with the output volume and quality of Chinese scholars in top journals, their influence in the international academic community is significantly lower.

As for why China has a high local bias, one possible answer is that Chinese scholars are willing to work together to cite each other, which is consistent with everyone's first impression that the Chinese academic community values ​​relationships. However, Qiu Shumin explained that there are two possible logics behind this question.

The first is the logic of sticking together, that is, Chinese scholars are too willing to cite "their own people" in the circle. The other logic is that foreign scholars cite too few papers of Chinese scholars. In this way, even if domestic scholars do not cite themselves excessively, the proportion of citations will show that the international mobility of research results is insufficient and the dissemination is concentrated within the country.

Which logic has a greater impact? Qiu Shumin added some data. 34.6% of the references of American scholars' top journal papers are from their own country. In contrast, the proportion of Chinese scientists' papers in the top journals is only 14.8%, which ranks only fifth in the world.

In other words, the proportion of papers cited by each Chinese scholar is not high, but because there are not many foreign scholars citing the academic achievements of Chinese scholars, the "citation pool" is relatively shallow, which makes China's self-citation ratio statistically abnormal. Although American scholars cited 34.6% of their own scholars in the references, considering that the world is citing American scholars' papers, this self-citation ratio has been diluted overall, and the self-citation phenomenon of American scholars does not seem so significant.

However, is it because of the quality of papers or other cultural and linguistic reasons that foreign scholars cite Chinese scholars less? If the quality of papers is similar, will the international citations of Chinese scholars' achievements be discounted? To answer these questions, the authors conducted another study.

Why do foreign scientists cite Chinese research less frequently?

This study focused on the field of chemistry. The authors first counted the PIs who had published at least one article in three top chemistry journals between 2000 and 2018, and then selected the top 1% of star scientists. After removing American scientists, they collected 751 resumes of non-American star scientists. Among them, 21%, or 156 people, were from mainland China, and the remaining 595 were from other countries and regions outside the United States. The authors collected all the articles of these star scientists after they became independent PIs.

The authors used a sophisticated matching method to control for potential differences in paper quality, matching across multiple variables, including journal, year of publication, number of authors, number of citations from non-US sources, year the scientist received his or her PhD, etc., in order to compare how papers by Chinese PIs and other non-US PIs were cited by a third party, the United States, when the quality was the same.

The study found that even after controlling for research quality, articles by Chinese PIs received, on average, 28% fewer citations from US scholars than articles by non-US PIs. This citation discount was seen in almost all years[5].

The team examined many explanations for this citation discount. First, this discount is not based on discrimination. Qiu Shumin said that she and her collaborators investigated the citations of 40 overseas (excluding the United States) and Hong Kong Chinese PIs and found that their articles were not treated differently. Compared with non-Chinese overseas scholars, these Chinese even received more citations from the American academic community.

Secondly, in terms of academic reputation, foreign scholars in some fields may be cautious about citing papers by Chinese authors due to the frequent retraction issues in China. However, the data does not support this hypothesis. Although the frequent retractions have been questioned, the citation rate of Chinese scientists' papers in fields with frequent retractions has not been significantly negatively affected.

In addition, there is speculation that the fields in which Chinese scholars focus are too niche, resulting in fewer international citations. However, the authors tested whether China's specialization in specific sub-fields would affect the citation rate of its research in the United States, and the results showed no significant effect.

The hypothesis that the author most supports is that American scholars do not have enough understanding of China's research results and that Chinese scholars are not sufficiently recognized (awareness) internationally.

Continuously deepening and focusing on research in their own fields can help Chinese scholars reduce the discount on citations. However, this is limited to those extremely focused PIs who are at the top of the distribution of sub-field importance, that is, those researchers who have reached the top level in their own professional fields.

A more obvious impact comes from academic networks. The study found that Chinese scientists who returned from the United States and were trained in the United States have higher citations in the United States. Although the citation discount cannot be completely eliminated, it can eliminate half of the citation discount. Another finding is that if one of the authors of the citing article has a Chinese name, then the frequency of citing Chinese scientific research results is equivalent to citing scientific research results from other countries, and there will be no significant reduction in the number of citations.

This may mean that Chinese researchers have more opportunities to interact with Chinese researchers in the United States, and such exchanges enhance Chinese scholars' understanding of their academic achievements. However, interaction with the wider American research community does not seem to be sufficient. The limitations of this academic network may limit the recognition of Chinese scholars in the international academic community.

"Returned scientists and overseas Chinese have played a very important role in the spread of Chinese academic knowledge. Chinese scientists returning from the United States will be less affected by negative factors. On the other hand, Chinese scholars working overseas have also promoted the spread and flow of Chinese academic knowledge outward," said Qiu Shumin.

But after the 2018 China Action Plan, the academic network between China, which is bridged by overseas Chinese and returnee scientists, has become more fragile.

Qiu Shumin conducted a study on Chinese co-authors of AI papers and found that from 2018 to 2021, although the proportion of cooperation between the United States and China in the field of AI declined, the proportion of cooperation with OECD countries and other countries increased, maintaining the proportion of international cooperation in the entire field. During the same period, China's cooperation with the United States, OECD countries and other countries declined across the board. The tense relationship between China and the United States in scientific cooperation has brought more uncertainty to the international influence of Chinese academic achievements.

Changes in Chinese paper collaborators in the field of AI. Provided by the author

Today, as the number of papers published in China and the number of high-quality papers among them are growing rapidly, the relatively insufficient international influence should be the focus of the academic community's next attention.

Qiu Shumin explained to The Intellectuals that from the perspective of innovation economics, the value of knowledge comes from its accumulation and diffusion. China has indeed invested a lot of manpower and funds in science, and has also achieved corresponding outputs. However, it is not enough to simply produce a large amount of knowledge. Knowledge needs to be diffused to exert its economic and academic value.

During a discussion, Qiu Shumin's collaborator once said: "China has many top scientists, and both input and output in science are high. But who is standing on the shoulders of Chinese scholars to do research? If a lot of knowledge is produced but no one uses it, this is also a problem."

She and her collaborators chose to conduct these studies in order to explore to what extent this knowledge has spread to the global scientific community and to what extent it has pushed the frontiers of human cognition forward, against the backdrop of the rapid increase in the total number of papers published in China and the number of high-quality papers among them in recent years.

"Now that the number of Chinese papers has been reached and the quality has also improved, what we should focus on next is whether your research has gained corresponding influence and voice? How much contribution has it made to the development of the entire global scientific community and to the advancement of the frontiers of human cognition?" said Qiu Shumin.

references:

[1] Institute of Scientific and Technological Information. (2023). China Science and Technology Paper Statistical Report 2023.

[2] Liu Shiyao. (2017, November 1). China's ranking of international paper citations rises to second place in the world. People's Daily Online

[3](2023, August 9). Chinese scientific research papers continue to win the triple crown, with the most citations in their own country. Nikkei Chinese website

[4]Qiu, S., Steinwender, C., & Azoulay, P. (2024).Paper Tiger? Chinese Science and Home Bias in Citations(No. w32468). National Bureau of Economic Research.

[5]Qiu, S., Steinwender, C., & Azoulay, P. (2022).Who stands on the shoulders of Chinese (scientific) giants? Evidence from chemistry(No. w30772). National Bureau of Economic Research.