Rethinking the author name ambiguity problem and beyond: The case of the Chinese context

ABSTRACT

The perennial problem of author name ambiguity has attracted increasing attention in the academic community. Drawing on the literature, this article first highlights the pervasiveness of the problem and discusses its adverse consequences. It then analyzes the behavioral causes of the problem in the Chinese context and attributes them to personal, cultural, and institutional factors. Informed by this analysis and recognizing ORCID as a promising solution, we propose an ORCID-based “Prevention plus Cure” campaign against author name ambiguity. The prevention objective relies on researchers’ consistent use of ORCID, while the cure objective involves retrospectively integrating ORCIDs into backfile publications. We also outline the responsibilities of various stakeholders to ensure the success of the campaign. Furthermore, we argue that universal adoption of ORCID can help curb authorship-related misconduct, discern predatory journals and publishers, and track researchers’ undesirable records of academic publishing. We then analyze the current status of ORCID adoption in China, identify potential challenges, propose tentative solutions to address them, and highlight ORCID as a tool that can be utilized to empower China’s combat against research misconduct. In conclusion, we emphasize the importance of conducting empirical research to inform more effective promotion of ORCID adoption in China.

KEYWORDS:

• Author name ambiguity
• Chinese names
• author name disambiguation
• ORCID
• retraction

Introduction

The author name ambiguity problem is classified by Shin et al. (2014) into two categories: “the namesake problem,” which pertains to identical names, and “the heteronymous name problem” (15), which relates to the use of different names by the same researcher. Identified with not only Chinese researchers but also their international colleagues with names in many other languages (Castillo 2009; Strotmann and Zhao 2012), the author name ambiguity problem has been attracting steady scholarly attention since the advent of digitalized academic publishing in the mid-1990s (Sanyal, Bhowmick, and Das 2021; Smalheiser and Torvik 2009). The globalization of science communication, the digitalization of academic publishing, and the need for efficient and accurate author-centric retrieval of bibliographic data necessitates author name disambiguation in the context of all languages. On the one hand, the large number of Chinese researchers have contributed considerably to enriching the scientific literature (Adams et al. 2023; To and Yu 2020; Woolston 2023). On the other hand, the author name ambiguity problem with Chinese researchers has posed a serious challenge to effective and efficient author-centric retrieval of bibliographic information (Harzing 2015; Sun and Zhou 2002; Tang and Walsh 2010). Although considerable progress has been made in trying to solve the problem over the past two decades, we are not there yet (Sanyal, Bhowmick, and Das 2021). Against this backdrop, it is now high time to rethink the author name ambiguity problem in the Chinese context and beyond. This article aims to heighten the awareness, in not only Chinese researchers but also the rest of the science community, about the gravity of the author name ambiguity problem and the importance of their joint efforts to solve the problem.

The pervasiveness of the problem and its adverse consequences

The author name ambiguity problem is wittingly summarized by Granshaw (2019) as “the issue with Wang and Smith” (236) because Wang is the most popular surname in China and Smith in five English-speaking countries (i.e., Australia, Canada, New Zealand, the UK, and the USA). Although Smith is a common English surname, it is “only the 123rd most common surname in the world” (Granshaw 2019, 239), outnumbered by quite a few Chinese surnames. As estimated by China’s Ministry of Public Security, round 85% of China’s population (i.e., more than 1.1 billion people) shared 129 surnames (Qiu 2008). Drawing on the Forebears database of names,¹ Granshaw (2019) found that there were 108, 105, 98, 75, 74 millions of Chinese surnamed Wang, Li, Zhang, Chen, and Liu, respectively and that approximately 40% of Vietnamese population were surnamed Nguyen. Similarly, nearly half of Korea’s population were surnamed Kim, Lee, and Park (Harzing 2015). The phenomenon of identical surnames for large proportions of the national populations in these countries suggests that the author name ambiguity problem is likely to be highly pervasive among academics from the countries.

In 2011, Y. Wang was identified as the most productive researcher name, which had appeared in the bylines of 3,926 academic publications, with an incredible publication rate of almost 11 papers per day (Butler 2012). Similarly, a search in the Web of Science revealed that Wang, Y had authored 1,116 publications in nanotechnology between 2000 and 2007 (Tang and Walsh 2010). Harzing’s (2015) investigation uncovered more myths about Y Wang’s research productivity. Specifically, Y Wang had managed to author around 30,000 articles between 1 January 2005 and 28 February 2015, with a publication rate of almost 9 per day. Y Wang had even published more than 100 papers in each of 73 distinct research areas and was affiliated with more than 500 universities in nearly 100 different countries. Further examination revealed that the initial Y referred to 6 different Pinyin transliterations of Chinese given names (i.e., Yang, Yi, Yuan, Ying, Yan, and Yu) in the last 10 papers by Y Wang. To zoom in on Yang Wang, the name authored 2,500 papers, with a weekly publication rate of 5, and was affiliated to more than 100 universities and even many different appointments within the same university. Similarly, Qiu’s (2008) search for Wang X in PubMed returned 8,904 entries, and Torvik and Smalheiser (2009) came to realize that approximately two thirds of author names in PubMed were ambiguous. A search on the ORCID website² on 7 December 2023 showed that there were 281,819 instances of Wang, including 2, 863 instances of Wang Wei, and 18,427 occurrences of Smith, including 192 occurrences of John Smith.

The author name ambiguity problem by nature makes it extremely challenging, if not impossible, to conduct efficient author-centric retrieval of bibliographic data. This adverse consequence affects various stakeholders who need a complete and accurate research profile of individual researchers for different purposes (see Smalheiser and Torvik 2009). Without efficient and accurate identification of target researchers, it is impossible to conduct reliable assessment of research productivity and impact of individual researchers (Harzing 2015; Strotmann and Zhao 2012). It is also time-consuming for researchers to identify potential collaborators whose full or abbreviated names happen to be shared with too many peer researchers. Similarly, it is a time-consuming process involving algorithms and manual data curation to compile accurate lists of Highly Cited Researchers and the World’s Top 2% Scientists. Careless research institutions may end up being deceived by researchers who fake their research productivity with a publication list padded with their namesakes’ publications (see a case of Tsinghua University as revealed by Qiu 2008). Journals and research funding agencies may be misinformed by a stellar list of publications contributed by multiple namesakes and consequently end up mis-identifying reviewers for manuscripts and grant proposals. For instance, the second author of this article, whose expertise lies in applied linguistics, was once misidentified as a namesake scholar and invited by a European national fund administration to review a grant proposal on e-government.

The causes of author name ambiguity in the Chinese context

Before examining the causes of the author name ambiguity problem in the Chinese context, it is necessary to present some facts about Chinese personal names. First, a Chinese personal name appearing in Chinese characters is made up of a surname and a given name, with the former being followed by the latter. Second, the majority of Chinese personal names consist of a one-character (monosyllabic) surname and a two-character (disyllabic) given name (Du 2007). Third, according to a recent national survey conducted by the Household Management Research Center of the Ministry of Public Security (HMRCoMPS, 2021), only 100 or so of approximately 1,600 Chinese surnames are frequently used, with the most popular five combined (i.e., Wang [王], Li [李], Zhang [张], Liu [刘], and Chen [陈]) accounting for 30.8% of the registered national population of China. Fourth, only 60 disyllabic surnames were included in The Hundred Surnames, and fewer than 10 remain in use (Dai 2006). The Chinese government encourages the naming of newborns with a disyllabic surname (double-barreled surname) composed of their parents’ two monosyllabic surnames to increase unique names (Du 2007). Fifth, although a large number of Chinese characters can be used as given names, a limited number of them are preferred, with differences in gender and changes over time (HMRCoMPS, 2021). Sixth, the Hanyu Pinyin (hereafter referred to as Pinyin) system was officially introduced by the State Council of China (1978) to transliterate Chinese personal and place names and has been adopted by the International Organization for Standardization (ISO) since 1982. Seventh, the tone-marking diacritics are omitted from Pinyin transliterations of Chinese personal names (State Council of China, 1978).

The author name ambiguity problem is manifested with four types of namesake in the context of the Chinese language. The predominant use of a limited number of Chinese surnames (Du 2007; He 2003) and nation-wide preferences for a small number of Chinese characters as given names (HMRCoMPS, 2021) result in a huge number of personal names identical both in Chinese characters and in Pinyin, which are identified as genuine namesakes in this article. As Mandarin Chinese is rich in homophones (Kałużyńska 2015; Liu, Shu, and Li 2007; ISO 2015), many Chinese personal names are homophonic namesakes, which are different in Chinese characters but identical in pronunciation and thus in Pinyin. For instance, 6 different surnames in Chinese characters (i.e., 计纪季蓟暨冀) are all pronounced as Jì (Dai 2006), and at least 9 different Chinese characters that may be picked as a given name (i.e., 韦伟玮炜纬苇帷闱嵬) are all pronounced as Wěi. The practice of not including tone markers in Pinyin transliterations of Chinese personal names can lead to homographic namesakes, whose corresponding names in Chinese characters are different. For instance, Hu Guangwei is a shared Pinyin transliteration of many possible different Chinese personal names if their tone markers are removed, such as Hú Guāngwěi, Hú Guǎngwéi, Hù Guāngwěi, and Hù Guǎngwèi, each of which may refer to many different personal names in Chinese characters. Initial-based namesakes may occur when Pinyin transliterations of personal names are presented as a combination of a surname and initials of a given name. For instance, Hu G. may refer to Pinyin transliterations of many different personal names, such as Hu Guang, Hu Guangwei, Hu Gong, Hu Gongxun, Hu Ge, and Hu Gehuan, each of which may refer to many different personal names in Chinese characters.

The ambiguity problem with Pinyin transliterations of Chinese personal names can be further complicated by the ordering of given names and surnames, the number of their syllabi, and the (dis)connection of disyllabic given names and surnames. Despite the adoption of the surname-first order (State Council of China, 1978), the surname-last order has been usually adopted proactively or passively when Chinese researchers publish in English-medium venues. Changing the order of the monosyllabic given name and the monosyllabic surname may make little, if any, difference to the possibility of a personal name having namesakes, as illustrated by Qiu (2008) with the examples of Wei Yang (魏洋) and Yang Wei (杨卫). In cases of a combination of a monosyllabic surname and a disyllabic given name, the two syllabi of the given name may be either hyphenated or separated. Separating a given name’s two syllabi may lead to the second syllable being misidentified as a monosyllabic surname. For instance, Xu Shaoxiong, when presented as Xu Shao Xiong, may be misidentified as Xiong Shao Xu and/or Xiong Shaoxu, which may refer to many different personal names in Chinese characters. Similarly, the hyphenated two syllabi of a given name may end up being misidentified as a disyllabic surname. For example, Wan Ling-Hu may be misidentified as Ling-Hu Wan, with the surname in the former being Wan (万) and in the latter Ling-Hu (令狐). Considering that there are just a few dozens of traditional disyllabic Chinese surnames (Qiu 2008), the possibility of mixing up disyllabic surnames with disyllabic given names is probably low for Chinese-knowing readers.

It should be noted that there are institutional inconsistencies in hyphenating non-monosyllabic given names and capitalizing surnames in Pinyin. The Pinyin system allows the capitalization of the first letter of surnames but disallows the hyphenation of polysyllabic given names (State Council of China, 1978). In contrast, the data norm for retrieval and evaluation of Chinese academic journals (hereafter referred to as the CAJ data norm) issued by the Editorial Committee of Chinese Academic Journals (ECCAJ 1999), as well as its revised version (ECCAJ 2006), mandates full capitalization of surnames and hyphenation of non-monosyllabic given names and double-barreled surnames. Such a requirement of hyphenation is not always followed in practice. Notably, the CAJ data norm is only applicable to English translations of publication by-lines in Chinese academic journals. As a result, Chinese researchers who have published in both Chinese and English journals may not hyphenate their non-monosyllabic given names in Pinyin when publishing in English-medium international journals.

Although the surname-first order is officially endorsed and promoted by ISO as an international standard for Chinese personal names in Pinyin, Chinese researchers have to apply the surname-last order most of the time when publishing in international outlets. As English has been a lingua franca for modern science communication (Mauranen 2003; Tardy 2004), most internationally well-recognized academic journals and bibliographic databases, which are owned and operated by publishers in English-speaking countries, follow the Anglo convention of placing surnames after given names and allowing the use of initials for given names. Additionally, bibliographic databases may index an additional set of author names in the form of surnames plus initials of given names even though given names are fully spelled in publication by-lines. Chinese researchers with educational and research experiences in Western countries tend to anglicize their surname position. Notably, it is a common practice within the Chinese communities in Hong Kong and Singapore that an anglicized given name precedes a transliterated Chinese name following the surname-first order (e.g., Tony Leung Chiu-wai and Lawrence Wong Shyun Tsai).

Since researchers do not have to use their legal names for science communication, some Chinese researchers with educational and/or research experiences in Western countries may adopt a Western given name and place it in front of their Chinese name following the surname-last order to form a hybrid new name for academic purposes (e.g., Andy Xuesong Gao). Some researchers may parenthesize their Western given name and place it between their Chinese given name and surname. For example, the first author of this article used to appear as Shaoxiong (Brian) Xu in his first six publications, in which the parenthesized English name was intended to better distinguish him. Unexpectedly, his hybrid name tends to appear on reference lists, without the initial of his Western given name, very likely because of the parentheses. As a result, he decided to remove the unhelpful parentheses and use the new hybrid name (i.e., Shaoxiong Brian Xu) throughout the rest of his academic career. Such a decision is supported by Harzing’s (2015) finding that researchers cited with more initials are less likely to be misidentified. A well-established scholar in the field of applied linguistics has also adopted more than one hybrid name, publishing as both Andy Xuesong Gao (e.g., Zheng and Gao 2016) and Xuesong (Andy) Gao (e.g., Gao and Zheng 2018). The examples of Andy and Brian clearly indicate that researchers’ inconsistence in using their names for scholarly publication is a cause of the author name ambiguity problem.

Apart from individual researchers’ personal choice, culture can also play a role in naming practices. Researchers may change their surnames to follow a local culture or tradition. For instance, married women in Hong Kong may place their husbands’ surname in front of their maiden name to form a two-character new surname. In the case of Lam Cheng Yuet-ngor, the former Chief Executive of Hong Kong, Lam is her husband’s surname and Cheng her maiden surname. Some female researchers married to Westerners may choose to drop their original Chinese surname and adopt their spouses’ Western surname. We know at least one example of this kind but refrain from sharing it for the sake of privacy. A researcher of any sex in a relationship may adopt their partners’ surname, permanently or temporally, but retain or discard their original surname.

Researchers may adopt different transliteration systems (e.g., Pinyin and Wade-Giles) to Romanize their personal names, depending on the culture which they were born into and/or brought up in. While the adoption of the Pinyin system in transliterating personal names has been mandated in China since 1978 (State Council of China), the Wade-Giles system and its variants are adopted within the Chinese community outside of China, where languages other than Mandarin Chinese, such as Cantonese, Hakka, Hokkien, Teochew and Hainanese, are spoken. Researchers in China who published internationally in English both prior to and after the official introduction to the Pinyin system may have published under different names due to the adoption of different transliteration systems. Last but not least, even if a comprehensive set of regulations on presenting personal names in publication by-lines were introduced and universally adopted for international science communication, honest error could be made by journals/publishers, bibliographic databases, and even researchers themselves in handling author names (Gasparyan et al. 2016). To conclude, the reasons for the author ambiguity problem can be personal, cultural, and institutional, as summarized in Table 1.

Table 1. Reasons for changing Pinyin transliterations of Chinese personal names.

Change to a Pinyin transliteration of a Chinese personal name	Reason for the change
	Institutional	Cultural	Personal
Anglicizing the surname position	√	√	√
Capitalizing the surname	√		√
Adopting a new surname		√	√
Adding a Western given name		√	√
Hyphenating a disyllabic given name	√		√
Separating a disyllabic given name		√	√
Replacing the given name with its initial(s)	√		√
Switching to a new transliteration system	√	√	√
Making an honest error in name handling	√		√

ORCID as a promising solution to author name ambiguity

Since Bagga and Baldwin’s (1998) call for author name disambiguation two and half decades ago, a plethora of algorithm-based techniques, which rely on various types of auxiliary information (e.g., affiliations, e-mail accounts, address, coauthors, and disciplines), have been proposed to tackle the author name ambiguity problem (for a comprehensive review, see Sanyal, Bhowmick, and Das 2021). However, without manual data curation, none of them can ensure complete data accuracy due to the lack or incompleteness of auxiliary information (Shin et al. 2014; Smalheiser and Torvik 2009) and the multi-authorship, multi-disciplinary and multi-institutional nature of academic publications (Smalheiser and Torvik 2009). In response, some bibliographic databases, such as Scopus and Web of Science, have established their own author identification system. However, the usefulness of those author identification systems is limited due to their database-specific applicability. Accordingly, there was an urgent need for a universal author identification system.

In 2009, the Open Researcher Contributor Identification (ORCID) Initiative was introduced as a long-desired ultimate solution to the challenging author name ambiguity problem (Haak et al. 2012). Three years later, ORCID registration was made available to individual researchers free of charge, and the yearly active ORCID researchers amounted to 8.09 million at the time of writing this article (ORCID n.d.-b). Given its increasing popularity, the ORCID solution is envisioned to solve the author name ambiguity problem. However, recent investigations identified “systematic variations in adoption and integration of ORCID into publication metadata” (Youtie et al. 2017, 437) and a relatively low frequency of ORCID use in publications archived by PubMed and MEDLINE (Boudry 2021). In particular, researchers from the USA and Asian countries, especially China, have adopted ORCID less frequently than their European counterparts have (Youtie et al. 2017). Therefore, we argue that it is now high time to launch an ORCID-based “Prevention plus Cure” campaign against the author ambiguity problem. While realization of the objective of prevention mandates every author’s consistent use of their ORCID, the objective of cure can be achieved by retrospectively integrating ORCID to publications, following the same logic as assigning DOIs to backfile publications.

The ORCID-based “Prevention plus Cure” campaign requires joint efforts and proactive participation of various stakeholders, as summarized in Table 2. Notably, there are four prerequisites for its success. First, every researcher should register for one ORCID account (preferably in one’s full name) for science communication and use it throughout their academic career. Second, journals/publishers should make it procedurally compulsory and technically convenient for authors to disclose ORCIDs in their own publications, as exemplified by Journal of Material Science (Carter and Blanford 2017) and many other journals. Third, the algorithm-based author name disambiguation techniques that are currently in use should be optimized to minimize data inaccuracy. Fourth, manual curation of author-related data and authoritative correction to inaccurate data should serve as a supplement to adopted algorithm-based author name disambiguation techniques. In addition to satisfying the above four prerequisites, stakeholders are also expected to do more as outlined below.

Table 2. Institutional stakeholder duties for the ORCID-based “Prevention plus Cure” campaign.

Goal	“Prevention plus Cure” campaign duty	IR	J/P	BD	RPO	RFA	ORCID
P	Use a name throughout one’s academic career	√
P	Provide sufficient author information in publications	√	√	√
P	Internalize the necessity of consistent ORCID use	√	√	√	√	√
P	Adopt an ORCID throughout one’s academic career	√
P	Publicize on one’s ORCID webpage a full record of publications	√
P	Promote ORCID use in one’s academic network	√
C	Report inaccuracy in author information	√
P	Require full ORCID disclosure at the submission stage		√
C	Optimize algorithm-based author name disambiguation techniques		√
C	Enable manual curation and correction of author-related data		√
P	Mandate disclosure of full names on reference lists		√
P	Encourage use of diacritics in transliterated names in by-lines		√
P	Allow use of full names in source languages as a supplement		√
P	Avoid replacing given names with initials		√	√
C	Retrospectively integrate ORCIDs to ORCID-absent publications		√
C	Update retrospectively adopted ORCIDs			√
P	Mandate ORCID use for academic publishing and reviewing		√		√	√
C	Assign an ORCID to deceased researchers without one						√
P+C	Establish a consortium for sharing resources and data		√	√	√	√	√

P= the Prevention objective; C = the Cure objective; P + C = the Prevention and Cure objectives; IR = individual researchers; J/P = journals/publishers; BD = bibliographic databases; RPO = research performing organizations; RFA = research funding agencies; ORCID = the ORCID organization.

Researchers should be consistent in using their name, be it a legal one or a pen name, for scholarly communication throughout their academic career. Those who have published in different names should stick to the current one and refrain from making further changes. Researchers with an ORCID should publicize on their ORCID webpages a full record of their own publications. By so doing, they can not only distinguish themselves from their namesakes but also provide reliable data for developing better algorithm-based author name disambiguation techniques. For the same purposes, researchers should always provide in their publications sufficient and accurate author-related information (i.e., author names, ORCID accounts, affiliations, e-mail accounts, and addresses) (Huang et al. 2014; Morillo, Santabárbara, and Aparicio 2013). Researchers are expected to recommend consistent use of ORCID to their coauthors, colleagues, and research students. They are also encouraged to report to journals/publishers and bibliographic databases any inaccuracy in author-related information. New entrants to the academic publishing system (e.g., research students and beginning researchers) should be made aware of the importance of a career-wide name for scholarly communication via academic supervision and/or institutional training. They are also advised to apply for an ORCID before submitting their maiden work for publication and to keep their ORCID accounts for subsequent publications.

Journals/publishers should consider modifying their regulations on author name disclosure. Specifically, they can mandate disclosure of full names of cited authors on reference lists. Making such a change is justified by the facts that full names are more immune than those with initials to name ambiguity (Harzing 2015) and that electronic academic publishing is free from the constraint of publishing space. Journals/publishers can also advocate use of tone-marking diacritics of author names transliterated from a language characterized with tone variations. Additionally, journals/publishers can consider allowing authors to provide in by-lines both their transliterated names and those in their mother tongues, with the latter being parenthesized and following the former, as American Astronomical Society (n.d.) journals do. Since these three suggested changes can only reduce the possibility of author name ambiguity, it is more desirable that journals/publishers focus on promoting the use of ORCID as a promising solution to the author name ambiguity problem. To this end, journals/publishers should make it mandatory for researchers to use ORCID in accepted manuscripts. Moreover, they should develop and follow a workflow that enables retrospective integration of ORCIDs into extant publications without such identifiers. Ideally, such a workflow covers all possible scenarios of ORCID absence (and maybe inconsistence). For instance, if a deceased author did not apply ORCID to some of his/her publications, the same ORCID should be integrated into all those publications posthumously.

Bibliographic databases, such as Web of Science, Scopus, and PubMed, should strictly apply the principle of modification-free data coverage in providing author-related information as part of metadata of all their indexed publications. In other words, they should avoid replacing authors’ full given names with initials and update ORCID integrated into publications retrospectively. The ORCID organization is expected to consider assigning an ORCID to every deceased researcher who did not have one, and then those posthumously assigned ORCID can be shared with and integrated retrospectively by journals/publishers into all the publications of deceased researchers. To endorse and promote the ORCID-based solution, other stakeholders (e.g., research institutions, research funding agencies, and conference organizers) can make it an institutional requirement for researchers to submit an ORCID-generated researcher profile. Last but not least, it is desirable that a consortium of various stakeholders (i.e., journals/publishers, bibliographic databases, research-performing organizations, funding agencies, and the ORCID organization) can be established to share author name disambiguation algorithms and data as part of orchestrated efforts to conduct the ORCID-based “Prevention plus Cure” campaign.

ORCID as a tool against misconduct in academic publishing

Our above discussion on the author name ambiguity problem was based on the assumption that every researcher publishes ethically and responsibly. Unfortunately, such an assumption is challenged by the pervasive phenomenon of ghost authorship and gift authorship (Flanagin et al. 1998). These two types of unethical authorship involve people taking up a position in a by-line despite a lack of (substantial) contribution to the published research or the process of writing it up. An alarming case of ghost authorship, as exposed by Perron, Hiltz-Perron, and Victor (2021), is that the publication is but a paper mill product derived from no research work at all. An exceptional scenario of gift authorship is imposed gift authorship, in which the authorship receiver remains unaware of the existence of the authorship giver and of being listed as a coauthor until the publication ends up being retracted due to misconduct (Marcus 2021). The worst scenario possible is a mixture of ghost authorship and imposed gift authorship, where an innocent researcher is unknowingly listed by an ill-intended researcher as a coauthor of an AI-generated paper mill article which ends up being retracted for misconduct, as revealed by Oransky (2021). Apparently, both ghost and gift authorship inflate researcher visibility through undeserved publications, intentionally or unintentionally, which can exacerbate the author name ambiguity problem when ghost and gift authorship receivers have any of the four types of namesakes mentioned earlier.

Researchers who have any of the four types of namesakes in academia may reap two windfall-like benefits at the cost of their namesake researchers. They can not only enlarge their oeuvre by amalgamating their namesakes’ publications with their own (Harzing 2015) but also impute their own undesirable publishing history (e.g., corrections, expressions of concern, and retractions) to their namesakes. These two problems can be effectively addressed through the universal adoption of ORCID in academic publishing. Disappointingly, over a decade after the launch of the ORCID initiative, it is still common that journals/publishers make it optional for corresponding authors to provide their ORCIDs at the stage of submission and add their coauthors’ only after article acceptance. If ORCID disclosure of all coauthors of a manuscript at the submission stage is made mandatory by journals/publishers, all the coauthors will be notified through their ORCID-linked e-mail account to confirm their authorship of the submitted manuscript and their consent to the submission. Such a workflow can not only reduce the risk of imposed gift authorship but also help to distinguish legitimate journals/publishers from predatory ones.

As it takes resources to ensure that every author in a publication by-line is attached to a unique ORCID, predatory journals/publishers, which are inherently profit-prioritizing, may ignore to make such an investment. If they happen to adopt ORCID, it will make their clients’ record of publishing with predatory journals/publishers easily retrievable, which may discourage their potential clients. Therefore, ignorance of ORCID can be seen as an indicator of predatoriness of journals/publishers. Fortunately, universal ORCID adoption will make it possible to systematically track individual researchers’ undesirable history of academic publishing. Specifically, it is envisioned that an ORCID-based author search in the Retraction Watch Database in the future will be returned with a full list of a specific researcher’s corrections, expressions of concern, and retractions, if any. To make better use of ORCID as an effective deterrent, if not a weapon, against various unethical publishing practices (e.g., paper mill publications, imposed gift authorship, and predatory publishing), research institutions and research funding agencies can consider requiring researchers to submit ORCID-adopting publications for research assessment (Haak, Meadows, and Brown 2018) and grant application (Notay 2018), as argued earlier on.

ORCID adoption in China

In this section, we first examine the current state of ORCID adoption in China, providing an overview of its integration within the Chinese science community. We then highlight the primary challenges that could hinder the effective implementation of ORCID in China. Following this, we offer a set of recommended solutions to address these challenges, aiming to enhance the adoption and utility of ORCID. We conclude by exploring how ORCID can be leveraged to strengthen China’s efforts to combat research misconduct, the main cause of a rocketing number of publication retractions.

The current state

According to our literature search on the China National Knowledge Infrastructure (CNKI), the ORCID initiative was first introduced to the Chinese science community in 2010. In 2014, the National Science Library of the Chinese Academy of Sciences launched the iAuthor platform,³ providing a convenient way for Chinese researchers to register for an ORCID account (Haak 2014). However, at the time of writing this article, only 69,670 researchers have registered for an ORCID account via the iAuthor platform. According to the ORCID statistics as of late January 2024, the number of Chinese researchers with active ORCID accounts was approximately 388,973, accounting for approximately 4.82% of the total of 8.07 million ORCID account holders worldwide (ORCID n.d.-b). These statistics indicate that ORCID has not been widely adopted in China,⁴ despite calls from many Chinese scholars for its adoption (Bai 2015; Huang, Yu, and Zhang 2017; Li, Xu, and Shi 2016; Ma and Huang 2016; Sun et al. 2015; Wang, Huang, and Liu 2015).

Notably, out of 1,356 ORCID member organizations from 59 countries, only 6 were based in China, accounting for less than 0.45% (ORCID, n.d.-a).⁵ This inactive involvement of Chinese organizations may have contributed to the low level of ORCID adoption among Chinese researchers. Notably, data obtained in early October 2020 revealed that approximately 177,700 researchers from 42 universities on the Double First-Class Construction scheme (hereafter referred to as Double First-Class universities) had registered for an ORCID account (Dou, Zhang, and Zhang 2022). This number accounted for over 45% of the active ORCID accounts as of late January 2024, indicating a significant representation of Double First-Class University researchers within the Chinese community of ORCID users.

Possible challenges

The participation of various stakeholders, such as journals/publishers, bibliographic databases, funders, research-performing organizations, and individual Chinese researchers is crucial for promoting ORCID adoption in the Chinese science community. However, not all manuscript handling systems adopted in China have integrated ORCID. It was reported that Chinese academic journals are required to pay for activation of the ORCID functionality embedded into their manuscript handling systems (Pan et al. 2020). This poses a challenge for ORCID adoption, as the majority of Chinese academic journals are operated independently by universities with limited financial resources and, therefore, may not be motivated to adopt ORCID. Furthermore, even among the Chinese academic journals that have taken the lead in adopting ORCID, many do not provide direct access to authors’ ORCID webpages, requiring readers to manually enter a 16-digit number into the ORCID website. This inconvenience not only hinders the visibility of ORCID but also underrepresent its utility. As a result of the low ORCID adoption by Chinese academic journals, none of the major Chinese bibliographic/citation databases, such as CNKI, Wanfang Data, and VIP, has enabled author search via ORCID, in contrast to their international counterparts like Scopus and Web of Science.

Since the ORCID record schema can provide a common information core required by research funders worldwide (Brown and Demeranville 2018), it is expected that Chinese research funders will eventually join their international counterparts (Notay 2018) in mandating ORCID adoption by grant applicants. However, as of now, no Chinese research funder, at any level, has become an ORCID member organization, and there is no indication that any Chinese research funder requires applicants to disclose their ORCID accounts. This lack of endorsement from their research funders may discourage some Chinese researchers from registering for an ORCID account. Despite the active investment of prominent Chinese research-performing organizations, such as Tsinghua University⁶ and Lanzhou University,⁷ in their development of institutional repositories, the adoption rate of ORCID within these repositories remains low. This is evident from the limited disclosure of ORCID accounts on individual researchers’ personal webpages. Notably, the Confederation of China Academic Institutional Repository⁸ does not allow ORCID-based author search on its database. This lack of ORCID adoption by Chinese academic institutional repositories can be attributed to ORCID’s narrow focus on research-related activities, which does not cater to the broader administrative needs of Chinese research-performing organizations. One area where ORCID falls short is in facilitating the administration of data on teaching performance, which is a crucial component of Chinese universities’ annual Key Performance Indicator exercises. Unfortunately, ORCID does not currently support teaching performance metrics. Their expectations of the all-in-one functionality of ORCID may hinder the adoption of ORCID among Chinese research-performing organizations.

Some Chinese researchers submitting manuscripts to Chinese academic journals may be unaware of the importance of ORCID, as well as reluctant to share their published research works (Pan et al. 2020; Wei 2015). Many other Chinese researchers may have already registered for an ORCID account and have to use it on international publishing platforms, where ORCID is often mandated for corresponding authors (ORCID, 2015). This could explain why Chinese researchers who primarily publish in Chinese academic journals may not actively use their ORCID accounts. They may not see the value in providing a complete and regularly updated record of their Chinese publications on the ORCID platform, especially if they have to manually upload them. This inconvenience may lead to their perception of the limited usefulness of ORCID. Furthermore, unlike publication records automatically synced from journals/publishers, manually added Chinese publications do not have institutional endorsement, which may reinforce Chinese researchers’ misperception of the limited usefulness of ORCID. Additionally, unlike the iAuthor platform, ORCID does not allow researchers to include a profile photo, making it impossible to generate a curriculum vitae with a profile photo directly from ORCID. This lack of a secondary functionality may discourage some researchers from fully embracing ORCID. The incomplete, inaccurate, and untimely nature of the data provided on many Chinese researchers’ ORCID webpages (Pan et al. 2020) may also have discouraged Chinese institutions from utilizing ORCID.

Security concerns have been identified in the Chinese literature as a major barrier to ORCID adoption in the Chinese science community, especially in regard to Chinese national security in science and technology. Some Chinese scholars (Dou, Zhang, and Zhang 2022) have raised concerns about the potential threat to the privacy of talents worldwide, as ORCID-based profiling of talents could enable precise monitoring and targeting of individual researchers. Informed by consultations with Chinese security experts, Chinese Journal of Cancer Research publicly announced its decision not to require ORCID disclosure from its authors (Dai 2021). The decision was made based on four main concerns: serious flaws in the identification of researchers, harm to academic freedom and the creation of a new monopoly in academic publishing, the potential threat to privacy from collecting and selling ORCID data, and the risk of arbitrary management of ORCID data (Dai 2021). In order to address these concerns and protect Chinese national security in science and technology, it is suggested that Chinese researchers should be encouraged to publish their research in domestic academic journals, and a nation-level system for managing the identities of Chinese researchers be established through legislation (Dai 2021).

Fear of potential negative consequences may deter some Chinese researchers from fully adopting ORCID. Complete adoption of ORCID would expose researchers’ undesirable publishing histories, including corrections, expressions of concern, or retractions. Having their full publication records displayed on their ORCID webpages might lead to reputational damage (Xu and Hu 2022). Furthermore, full adoption of ORCID would facilitate the efficient retrieval of individual researchers’ retracted publications, potentially resulting in various tangible sanctions, such as exclusion from research funding applications (Tang, Wang, and Hu 2023). To avoid these adverse effects, some researchers might intentionally create and use multiple ORCID accounts for different aspects of their science communication, thereby undermining ORCID’s primary function of linking individual researchers with all their research outputs.

Suggested solutions

Promoting the adoption of ORCID among Chinese researchers can be achieved by effectively addressing the security concerns of the Chinese science community, offering ORCID functionalities that cater to a wider range of needs of Chinese researchers and institutions, and highlighting successful cases of ORCID adoption in China. To ensure the data security of Chinese ORCID users, ORCID could explore the possibility of establishing a dedicated data server within China, similar to Apple’s establishment of a data center in Guizhou exclusively for storing the data of Chinese Apple users. This localized data storage approach could enhance trust and compliance with Chinese regulations while safeguarding the privacy and security of Chinese researchers’ data on the ORCID platform. Furthermore, it is recommended that ORCID further expand its utility by encompassing all academic-related data, such as administrative services, teaching, supervision, and social outreach. Ideally, the functionality and data coverage of ORCID should enable the automatic generation of comprehensive CVs for academics. This would provide a more holistic view of an individual’s academic contributions and facilitate streamlined documentation and presentation of their academic achievements.

Conducting case studies can provide valuable insights into the adoption of ORCID in select journals included in the Excellence Action Plan for China’s STM Journals. This plan, initiated by the China Association for Science and Technology with the aim of internationalizing Chinese STM journals, provides a strategic opportunity for promoting ORCID adoption. Additionally, conducting case studies on Chinese journals that have taken the lead in adopting ORCID, such as New Technology of Library and Information Service, Chinese Journal of Scientific and Technical Periodicals, Journal of Library Science in China, and Journal of Zhejiang University (Humanities and Social Sciences), can serve as compelling examples to showcase successful ORCID adoption. Furthermore, showcasing the adoption of ORCID in institutional repositories of some of the current 147 universities on the Double First-Class Scheme, such as Tsinghua University and Lanzhou University, can serve as compelling models in China. These universities, driven by their aspiration to improve global university rankings, may prioritize publishing in international journals over domestic ones. As a result, they are likely to demonstrate a higher motivation for adopting ORCID.

The Hong Kong Principles for assessing researchers highlights the importance of recognizing ORCID adoption as a commendable contribution to the open science movement (Moher et al. 2020). Individual researchers who are actively involved in academic publishing should take the lead in embracing ORCID and fostering a culture of ORCID-facilitated open science. Without such a culture, the Prevention objective of the “Prevention plus Cure” campaign cannot be achieved, thereby making the attainment of the Cure objective even more challenging. The establishment of a culture of ORCID-facilitated open science heavily relies on the commitment of individual researchers to honesty, accountability, and transparency in their academic activities.

It is important for ORCID to develop technical solutions that minimize the occurrence of duplicate ORCID accounts and maximize the visibility of full publication records for ORCID users. By implementing measures that deter individuals from gaming the academic publishing system with multiple ORCID accounts, ORCID can uphold the system’s integrity and reliability, thereby encouraging more institutions to adopt ORCID. In addition to technical solutions, it is vital to educate the science community about the significance of disclosing undesirable publishing histories. Instead of being viewed in a negative light, these disclosures should be acknowledged by the science community as compelling evidence of researchers’ dedication to honesty, accountability, and transparency after the fact. Transparently acknowledging and addressing mistakes or issues in the publishing process contributes to the overall integrity of scientific research and promotes responsible scholarly communication. As a result, researchers who proactively disclose their undesirable publishing histories can gain positive recognition from the science community, as advocated by Benoit’s (2015) image repair theory.

ORCID-empowered fight against research misconduct

It is recommended that ORCID establish records for researchers’ breaches of research integrity (Chen 2016), which can be utilized by research administrators in identifying cases of research misconduct and subsequently imposing sanctions. In other words, the ORCID adoption among researchers with a poor record of research integrity may result in not only reputational damage but also concrete consequences. For instance, in August 2022, the Chinese Ministry of Science and Technology, along with 21 other national governing bodies, introduced a comprehensive set of regulations on investigating and sanctioning violations of research integrity.⁹ These regulations aim to serve as a strong deterrent against research misconduct among Chinese researchers. However, for these regulations to be effective, it is essential to precisely and efficiently identify researchers with a record of publications retracted due to research misconduct. This is where the connectivity provided by ORCID, which links researchers to their complete publication records, can play a significant role.

A global success of the ORCID-based “Prevention plus Cure” campaign can facilitate the identification of researchers with a record of retractions due to research misconduct, making the process efficient and, ideally, effortless. Given that China has a larger number of retracted publications and a higher retraction rate compared to other countries (van Noorden 2023), it can benefit more from this initiative. To strengthen research integrity and combat research misconduct, China must impose appropriate sanctions on researchers with a record of retractions due to research misconduct. Universal ORCID adoption can enable efficient identification of repeat offenders with multiple retractions while avoiding the inadvertent penalization of their namesakes without a retraction record or with fewer retractions. By leveraging the ORCID capabilities, the “Prevention plus Cure” campaign can empower China’s fight against research misconduct.

Concluding remarks

Once fully implemented, ORCID can effectively address the issue of author name ambiguity, thereby enabling efficient researcher-centric retrieval of bibliometric data and thus facilitating the evaluation of research performance at the micro level of individual researchers. Researcher-specific evaluation of research performance should include retracted publications by treating them as a discrediting indicator of research performance, given their contamination of the scientific literature at the least, especially in cases of research misconduct. As evidenced by China’s recent national review of retractions (Mallapaty 2024), the Chinese authorities overseeing research integrity urgently need to swiftly track retracted publications and accurately identify the involved researchers. By endorsing and promoting participation in the ORCID-based “Prevention plus Cure” campaign, China can strengthen its combat against research misconduct, foster a culture of accountability, ensure the integrity of its scientific output, and ultimately protect its reputation in the global science community.

However, despite the necessity and the benefits of ORCID adoption, the Chinese science community may face various obstacles in fully embracing ORCID, such as limited resources, dissatisfaction about the current functionality of ORCID, national security concerns related to science and technology, and fear among misbehaving researchers of adverse consequences from ORCID adoption, as discussed above. Another factor influencing ORCID adoption in China may be a collective tendency to overlook breaches of research integrity. As of 2021, the Retraction Watch Database had documented 15,727 retracted publications affiliated to Chinese institutions, accounting for approximately 46% of the worldwide total of 34,223 retractions.¹⁰ In contrast, only 404 publications had been retracted from Chinese academic journals through retraction notices as of 2021 (Xu 2023). Despite Chinese academic journals publishing a significantly higher number of research papers by Chinese researchers compared to international journals, they had seen drastically fewer retractions. However, it can be reasonably inferred that if Chinese academic journals had adopted the retraction criteria followed by international journals, a similar, if not greater, number of retractions would have been expected. In light of this situation, if ORCID ends up being abandoned by the Chinese science community, it would become imperative for China to develop its own researcher identification system and mandate nationwide adoption of it. Such a national identification system would enable the effective identification of misbehaving Chinese researchers, providing a powerful tool to curb research misconduct more effectively. However, the contrasting behavior of Chinese researchers in international and domestic journals, as suggested by the significant difference in the number of retractions, raises the question of whether a robust identification system like ORCID or its Chinese equivalent would be truly considered necessary and welcomed in the Chinese science community.

Whether China fully embraces ORCID or decides to establish its own national researcher identification system for any reason, the international science community may not experience a significant impact as long as the “Prevention plus Cure” campaign, hopefully implemented outside of China, covers all the research outputs of Chinese researchers appearing on international publishing platforms. However, if China eventually comes up with a national researcher identification system that does not interoperate with ORCID, it could inconvenience Chinese researchers who publish internationally, as they would have to navigate between two different identification systems. The worst scenario would be that China’s development and implementation of its own researcher identification system could potentially isolate the Chinese science community from the international science community, impeding international collaboration and hindering its progress in science and technology. To promote effective ORCID adoption in China or inform the development and implementation of a Chinese national researcher identification system, it is crucial to conduct empirical research on the current state of ORCID adoption in China. The existing findings on ORCD adoption in China have predominantly relied on observational data and relevant literature (e.g., Pan et al. 2020), highlighting the need for empirical research to assess the extent to which various obstacles have impeded ORCID adoption in China. Therefore, future empirical research should analyze the current status of ORCID adoption in China, with a specific focus on key stakeholders, such as prominent Chinese academic journals, major bibliographic databases, and researchers in Double First-Class universities and national research institutions. Further research is also needed to explore the degree to which ORCID can unambiguously identify Chinese authors of retracted publications.

It is worth noting that it took 15 years for DOI to be officially recognized and widely adopted in China. Initially introduced to the Chinese science community in 2000, DOI was only included in China’s National Standards in 2015. Similarly, ORCID was first introduced as an initiative to the Chinese science community in 2010 and was adopted internationally in 2012. However, despite its introduction 12 years ago, the Chinese science community is still hesitant in fully embracing ORCID. It is possible that it may take even longer for the Chinese science community to fully recognize and adopt ORCID compared to DOI. Additionally, the introduction of Research Organization Registry (ROR) in 2015, which aims to distinguish every research organization in the world with a persistent identifier, further emphasizes the importance of persistent identifiers for digitalized international science communication. Surprisingly, at the time of writing this article, our search on CNKI yielded no literature introducing ROR. The three identification systems, namely ROR, ORCID, and DOI, work together to form the infrastructure for digitalized international science communication. It remains to be seen when and how the Chinese science community will adopt all the three identification systems for the benefits of both the Chinese and international science community.

Acknowledgments

We would like to express our sincere appreciation to the ORCID Support team for their prompt and helpful responses to our inquiries regarding ORCID. We would also like to acknowledge the valuable assistance provided by Dr. Huimei Zhang at Huanggang Normal University and Ms. Hongping Xu at Sanjiang University in investigating the ORCID adoption by Chinese academic journals. Furthermore, we are grateful to the three reviewers for their constructive critical feedback, which greatly contributed to the expansion and improvement of the original version. Any errors that remain in this article are solely our own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes

1. https://forebears.io

2. https://orcid.org/orcid-search/search

3. http://iauthor.cn/welcome/index

4. In this article, China refers to the mainland of China, excluding Hong Kong, Macau, and Taiwan.

5. The query was conducted on February 4, 2024.

6. https://thurid.lib.tsinghua.edu.cn

7. https://ir.lzu.edu.cn

8. http://chair.calis.edu.cn

9. https://www.most.gov.cn/xxgk/xinxifenlei/fdzdgknr/fgzc/gfxwj/gfxwj2022/202209/t20220907_182313.html

10. The query was conducted in the Retraction Watch Database on February 4, 2024.